Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: XAUI IO specs




This is confusing.  If Vo+ is VCM+400mV in logic 1 and VCM-400mV in logic 0 then the swing would be
800mV, but you say each side has a 400mV swing.  My understanding of everything that everyone else
has said is that the 400mV swing on each output is correct which would have

 - Logic 1 would defined as Vo+ at (VCM+200 mV) and V0- at (VCM-200mV)

 - Logic 0 would defined as Vo+ at (VCM-200 mV) and V0- at (VCM+200mV)

ghiasi wrote:

> Hi
>
> Several question have been raised about XAUI "Hari" specifications and what was some of the
> intentions:
>
> o. Based on my understanding and everyone who participated during conference call the Vp-p
> differential max defined as 800 mV.
>
>         - This means each side of the driver has a swing of 400 mV.
>
>         - Logic 1 would defined as Vo+ at (VCM+400 mV) and V0- at (VCM-400mV)
>
>         - Logic 0 would defined as Vo+ at (VCM-400 mV) and V0- at (VCM+400mV)
>
>         - If the driver output VDp-p <85 mV you got noise.
>
> VCM- Voltage Common Mode
> VDp-p Voltage Differential p-p
>
> o. During the time we defined Hari, we explicitly left out reference to VCM.  This was due to
> interoperablity of different logic voltage levels and technologies.  For example most of the
> existing SerDes today, it is not possible to connect the driver directly to the same part
> receiver.  At the same time there are some applications such as large switches which may be
> very desirable not to use Ac-coupling caps.  In some of these applications you may end up using
> several thousands of caps on a backplane.  In another standard I proposed the followings an is
> incorporated in to the specifications:
>
>         - If the transmitter does not provide VCM of 0.75 V, then must add AC caps
>         - If the receiver can not tolerate 0.75 V of VCM, then it must add AC caps.
>
> Overall the above definition of VCM would not burden the implementation as it is optional and
> only high applications need it.
>
> o. I have also had several comments on the drive amplitude. If we decide to define copper
> twin-ax cable of length 10-15 m, we would need to increase the amplitude somewhat.  IB drive
> amplitude are higher than Hari.  One option here might be:
>
>         - Drive hari levels pre-emphasis Off
>         - Drive IB levels turn pre-emphasis On
>
> I hope this answers some of your question.
>
> Thanks,
>
> Ali Ghiasi
> Sun Microsystems
> (650)786-3310
>
>
> > From: kdemsky@xxxxxxxxxx
> > X-Lotus-FromDomain: IBMUS
> > To: HSSG <stds-802-3-hssg@xxxxxxxx>
> > Date: Mon, 17 Apr 2000 21:25:03 -0500
> > Subject: Re: XAUI IO specs
> > Mime-Version: 1.0
> > Content-Disposition: inline
> > X-Resent-To: Multiple Recipients <stds-802-3-hssg@xxxxxxxxxxxxxxxxxx>
> > X-Listname: stds-802-3-hssg
> > X-Info: [Un]Subscribe requests to  majordomo@xxxxxxxxxxxxxxxxxx
> > X-Moderator-Address: stds-802-3-hssg-approval@xxxxxxxxxxxxxxxxxx
> >
> >
> > Rich,
> >
> > Since I am an electrical PHY guy, I have been lurking and waiting for the
> > I/O subject to surface.  It may be premature to talk about I/Os when the
> > overall solutions have not been nailed down yet, but since the subject has
> > been raised, I will offer clarification of the I/O issue at hand.
> >
> >
> > an excerpt from Rich's reply,
> >
> > >When the Output is set to a logic one, the positive differential signal,
> > >referenced to a 0 V ground, is a maximum of +400 mV and the negtive
> > differential
> > >signal, referenced to a 0 V ground, is a maximum of -400 mV.
> >
> > >When the Output is set to a logic zero, the positive differential signal,
> > >referenced to a 0 V ground, is a maximum of -400 mV and the negtive
> > differential
> > >signal, referenced to a 0 V ground, is a maximum of +400 mV.
> >
> > >I'd call this V diff peak-to-peak, but this I'm NOT an expert in this
> > area.
> >
> > >Best Regards,
> > >Rich
> >
> > The above attempt at clarification made me more confused.  No offense
> > intended, Rich.  :)  The way the above is worded I can still take it two
> > ways.  I think if I look at the way each signal is described above, and in
> > the logic one case positive output is +400 from the common mode, and the
> > negative output is -400 from the common mode, then we have an 800 mV
> > differential for logic one, and also folowing the next paragraph we have
> > -800 mV leaving Vdiff =1600 mV peak-to-peak.  I don't think this is the
> > case for several reasons.
> >
> > Perhaps the numbers came from or match the IEEE LVDS spec on page 10 of
> > IEEE Std 1596.3-1996.  The common mode is intended to be 1.2 Volts, and the
> > differential is NOT specified peak-to-peak, but merely differential.  |Vd|
> > has absolute value signs around it and therefore doesn't care about
> > polarity.  In this case, |Vd| max = 400mV.  The picture has the output
> > driver at a common mode of 1.2 Volts,  and each of 2 differential signals
> > swings single ended from 1.0 to 1.4 Volts.  In one case the difference is
> > +400 mV (positive peak), and when polarity switches the difference is
> > -400mV (negative peak).  The differential peak-to-peak then is +400 -
> > (-400) = 800 mV.
> >
> > Differential I/O Specs can be confusing.  Darn those factors of two!
> > Additionally, it is nice to know the common mode  (average of the two
> > signals).  If you truly mean LVDS, please reference IEEE Std, or
> > ANSI/TIA/EIA, or other LVDS compatible.  Or, in the absence of referring to
> > a spec, draw a picture of the signal or write 1000 words.
> >
> > I also realize that there are bigger fish to fry than I/O levels and it may
> > be premature to talk about I/O levels.    However, if the XAUI/XGXS
> > solutions are maturing, please offer clear I/O definitions.
> >
> > For happy reading, OIF (Optical Internetworking Forum) Document
> > #OIF99.102.5 has defined a common electrical interface between SONET framer
> > and serializer/deserializer parts for OC-192.  It looks to be a fairly
> > mature document for what HSSG calls the PCS/PMA interface.  This would be
> > the 16-bit serializer/deserializer and clock and data recovery spec.  Some
> > work could be leveraged from this document.
> >
> >
> >
> > Kevin Demsky
> > Mixed Signal and VLSI Development
> > IBM Corp.
> > 3605 Hwy 52 N
> > Dept. QXS Bldg. 050-2
> > Rochester, MN  55901
> >
> > Internal E-mail:  kdemsky@ibmusm07
> > External E-mail:  kdemsky@xxxxxxxxxx
> >
> > Phone:  507-253-5799
> > Fax:  507-253-4966
> >
> >
> > Rich Taborek <rtaborek@xxxxxxxxxxxxx> on 04/15/2000 05:05:52 PM
> >
> > Please respond to rtaborek@xxxxxxxxxxx
> >
> > To:   HSSG <stds-802-3-hssg@xxxxxxxx>
> > cc:
> > Subject:  Re: XAUI IO specs
> >
> >
> >
> >
> >
> > Mark,
> >
> > I'll have to defer to Mr. Ali Ghiasi and Mr. Richard Dugan on this issue to
> > properly resolve the nomenclature. My understanding of the parameters
> > listed is
> > as follows using Vo Dif(max) 800 mV as an example:
> >
> > When the Output is set to a logic one, the positive differential signal,
> > referenced to a 0 V ground, is a maximum of +400 mV and the negtive
> > differential
> > signal, referenced to a 0 V ground, is a maximum of -400 mV.
> >
> > When the Output is set to a logic zero, the positive differential signal,
> > referenced to a 0 V ground, is a maximum of -400 mV and the negtive
> > differential
> > signal, referenced to a 0 V ground, is a maximum of +400 mV.
> >
> > I'd call this V diff peak-to-peak, but this I'm NOT an expert in this area.
> >
> > Best Regards,
> > Rich
> >
> > --
> >
> > Mark Kerestes wrote:
> > >
> > > Rich, Ali
> > >  On slide 22 of the XAUI/XGXS Proposal from Albuquerque was listed :
> > > Vo Dif(max) 800mV
> > > Vo Dif(min) 500mV
> > > Vin Dif(max)  1000mV
> > > Vin Dif(min) 175mV
> > > I assume these are truely Vdif and NOT Vdif peak to peak? I have some
> > > confusion from looking back at the
> > > Nov 5 1999 "HARI the Electrical Interface" which seems to have a table
> > > that mixes Vdif and Vdif peak to peak
> > > showing PECL vs LVDS vs HARI side by side.
> > > Best regards,
> > > Mark
> >
> > -------------------------------------------------------
> > Richard Taborek Sr.                 Phone: 408-845-6102
> > Chief Technology Officer             Cell: 408-832-3957
> > nSerial Corporation                   Fax: 408-845-6114
> > 2500-5 Augustine Dr.        mailto:rtaborek@xxxxxxxxxxx
> > Santa Clara, CA 95054            http://www.nSerial.com
> >
> >
> >