From: Jim Freeman ([email protected])
Date: Thu Dec 02 1999 - 12:57:26 PST
Hi Steve,
You are correct. There are both differential and common mode issues when differential signals are employed. Most commonly, the differential impedances
are either ignored or deemed not significant. A proper stachup satisfies both of these criteria for proper signal transmission. It can be said that many
versions of this have been attenpted and a problem was not seen on the board, including those who basically ignored the differential considerations. This
probably means that the the performance required was not high enough to see a problem or the designer lucked out. Much can be said for lucking out but as
we reach higher performance systems, both the differential and common mode requirements of differential signalling will need to be met.
Jim Freeman
sweir wrote:
> Jim,
>
> I would just like to add to that in the case of the differential signaling
> the reflection plane still carries the majority of the reflection currents
> for each conductor. Because these currents are almost equal, at any
> distance from the pair the net current will appear to be very close to
> zero. However, at the signal transitions, the common mode reflection can
> be quite significant.
>
> Regards,
>
> Steve.
> At 03:50 PM 12/1/99 -0800, you wrote:
> >Hi Aubrey,
> >
> > In the case of a single signal travelling down a transmission line, the
> >other plane/conductor has an induced value that travels in the opposite
> >direction.
> > In the case of differential signalling, there are two signals
> >explicitly driven by drivers that force equal and opposite values to be
> >driven down the differential pair with a differential ground that is
> >exactly the average of the two signals.
> >
> > These are distinct cases because of the driving scheme.
> >
> >Jim Freeman
> >
> >
> >[email protected] wrote:
> >
> > > Since TEM and TM modes are mentioned on the same page as Differential
> > > signalling, I would like to request some clarity when Differential
> > > signalling is discussed.
> > > Why?
> > > Well, you can have Differential signalling when two separate signals are
> > > received by a differential amplifier or comparator even though the two
> > > signals may travel via two separate single ended microstrip transmission
> > > lines.
> > > Isn't it also Differential signalling when you have a single signal
> > > traveling down a balanced transmission line? IE, twin lead, co-planar
> > > structures. Hmmmm. Different effects, Different problems, Different
> > > solutions. Lots of different stuff...
> > >
> > > Aubrey Sparkman
> > > Signal Integrity
> > > [email protected]
> > > (512) 723-3592
> > >
> > > -----Original Message-----
> > > From: Jim Freeman [mailto:[email protected]]
> > > Sent: Wednesday, December 01, 1999 2:08 PM
> > > To: si-list
> > > Subject: [SI-LIST] : topics for discussion
> > >
> > > Hi All,
> > > Adrian talked about basic topics for discussion. Here are a few
> > > ideas.
> > >
> > > 1. TEM and TM modes in transmission lines or for EMC. What are the
> > > assumptions that drive one or the other?
> > >
> > > 2. Differential signalling - broadside coupling or edge coupling,
> > > differential mode ground and common mode ground, EMC implications for
> > > different types of coupling.
> > >
> > > 3. Methods for describing more accurately the behavior of transmission
> > > lines and return paths under changing frequency conditions.
> > > Incorporation of dielectric losses in the simulation and/or
> > > calculations. Incorporation of the harmonics of the digital signal with
> > > corresponding changes in impedance to better emulate dispersion of the
> > > digital signal that gives rise to jitter/skew.
> > >
> > > These are just a few of the more esoteric concepts that I would like to
> > > see discussed.
> > >
> > > Thanks
> > > Jim Freeman
> > >
> > > **** To unsubscribe from si-list: send e-mail to
> > > [email protected]. In the BODY of message put: UNSUBSCRIBE
> > > si-list, for more help, put HELP.
> > > si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list
> > > ****
> >
> >
> >**** To unsubscribe from si-list: send e-mail to
> >[email protected]. In the BODY of message put: UNSUBSCRIBE
> >si-list, for more help, put HELP.
> >si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list
> >****
>
> **** To unsubscribe from si-list: send e-mail to [email protected]. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list
> ****
**** To unsubscribe from si-list: send e-mail to [email protected]. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list
****
This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 11:39:04 PST