FreeLists / si-list / [SI-LIST] Re: AC Coupled Signals

["Loyer, Jeff" <jeff.loyer@xxxxxxxxx>]

A late-night epiphany has finally clued me in to reconciling the
physics.

I was doing my pulse response simulation with the source and load
impedances equal.  When I changed my source impedance to 8 ohms
(typical, but arbitrary) and left my load impedance at 50 ohms, the
pulse responses were dramatically different, depending on which side I
excited the system from.  I believe this is what you were alluding to,
though I didn't comprehend it as such.

This makes sense, and demonstrates how 2 systems can have IDENTICAL
insertion loss and yet have dramatically different pulse response/eye
diagrams.  Very interesting...

Thanks for sharing your insights.

Jeff Loyer


-----Original Message-----
From: Istvan.Novak@xxxxxxx [mailto:Istvan.Novak@xxxxxxx]=20
Sent: Monday, October 01, 2007 4:15 PM
To: Loyer, Jeff
Cc: si-list@xxxxxxxxxxxxx
Subject: Re: [SI-LIST] Re: AC Coupled Signals

Jeff,

My talented friend, Gustavo Blando, with whom we analyzed this situation
a few years back, reminded me that the voltage-transfer function
formulas can be found for instance in "MicroWave Transistor Amplifiers",
Guillermo Gonzales, Page 185.
You can also look at L. S. Dutta, T. Hillmann-Ruge, "Application of Ring
Oscillators to Characterise Transmission Lines in VLSI Circuits," IEEE
Transactions on Components, Packaging, and Manufacturing Technology -
Part B., Vol. 18, No. 4, November 1995, pp. 651-657. Note that you will
find various expressions; you may want to pick the Vout/Vsource
formulas, as opposed to Vout/Vin. You will see that the voltage transfer
ratio is, in general, different from S21.

Regards,
Istvan Novak
SUN Microsystems


Loyer, Jeff wrote:

>Hello Istvan,
>You're keying in to what was surprising to me.  For the asymmetric=20
>lossy network I simulated, I noted that the pulse response remained=20
>constant, regardless of direction of simulation.  Of course, S11 and=20
>S22 changed dramatically.
>
>The pulse response was not a derivation of the s-params; it was merely=20
>Vout (the circuit driven with an arbitrarily chosen 2V source) of a=20
>simulation of the actual network (3 lossy lines).
>
>So, it appears that the pulse response and insertion loss remained=20
>constant, even though the return loss varied dramatically.  I wasn't=20
>surprised to have the insertion loss stay constant, but couldn't have=20
>predicted whether the pulse reponse would also remain constant, though=20
>I suspected it would.  This implies the eye diagram would also be the=20
>same, regardless of direction of excitation.
>
>Having the pulse response remain constant appears contrary to your=20
>experience.  Any explanation?
>
>Thanks for your thoughts,
>Jeff Loyer
>
>
>-----Original Message-----
>From: Istvan.Novak@xxxxxxx [mailto:Istvan.Novak@xxxxxxx]=3D20
>Sent: Monday, October 01, 2007 4:46 AM
>To: Loyer, Jeff
>Cc: istvan.novak@xxxxxxxxxxx; si-list@xxxxxxxxxxxxx
>Subject: Re: [SI-LIST] Re: AC Coupled Signals
>
>Hi Jeff,
>
>Let me see if I can summarize this in simple physical terms. The=20
>scattering matrix deals with power values.  In a reciprocal  network,=20
>the 'through' power is the same regardless of the direction we go.  The

>node voltages, on the other hand, depend on the local impedance levels,

>which are related to S11, S22 and source and load reflection=20
>coefficients.=3D20
>
>So probably a more generalized condition is that we can shuffle around=20
>reciprocal building blocks inside a cascaded network, and the resulting

>eye diagram will stay the same as long as we do not change S11 and S22.
>
>Going back to your simulations: if you calculate the pulse response of=20
>your example circuits strictly from S21 of the network, I agree, it=20
>will not change, since S21 will not change as you move the reciprocal=20
>building blocks around.=3D20 Your eye diagram, however, should be =
usually

>calculated as Vout/Vsource (and for sake of simplicity, we usually=20
>assume linear driver and receiver impedances, where their linearity=20
>does not change this argument). This voltage transfer ratio can be=20
>analytically calculated if you wish, from the S-parameter flow graph. =20
>The resulting formula is symmetrical in S12 and S21, but asymmetrical=20
>in S11, S22 as well as source and load reflection coefficients.=3D20 If =

>you simulate the transfer response or the pulse response of=20
>Vout/Vsource in HSPICE, you should see the change as soon as the=20
>electrical symmetry is changed.
>
>Regards,
>
>Istvan Novak
>SUN Microsystems
>
>
>
>
>
>
>Loyer, Jeff wrote:
>
> =20
>
>>Hi Istvan,
>>Could you point me towards more information on the "voltage=20
>>transfer=3D20 ratio", and the difference between it and s21/s12?  When =
I

>>did the=3D20 pulse response of a system where S21 =3D3D3D S12 (though =
S11=20
>>and S22 were =3D
>>   =20
>>
>
> =20
>
>>very different, and the reflections were significant), it came =
out=3D20=20
>>equal (p21 =3D3D3D p12, see my posting of 9/15).  Thus, I think the=20
>>eye=3D20 diagram will =3D3D come out equal if S21 =3D3D3D S12.  =
=3D3D20
>>
>>Thanks,
>>Jeff Loyer
>>
>>
>>-----Original Message-----
>>From: si-list-bounce@xxxxxxxxxxxxx=3D20
>>[mailto:si-list-bounce@xxxxxxxxxxxxx]
>>On Behalf Of Istvan Novak
>>Sent: Saturday, September 29, 2007 12:38 AM
>>To: Chris.Cheng@xxxxxxxx
>>Cc: lifeatthesharpend; signalintegrity@xxxxxxxxxxx;=20
>>ron@xxxxxxxxxxx;=3D20 si-list@xxxxxxxxxxxxx
>>Subject: [SI-LIST] Re: AC Coupled Signals
>>
>>Chris and All,
>>
>>I think the source of misunderstandigs might stem from the fact that=20
>>in
>>   =20
>>
>
> =20
>
>>linear, time-invariant, reciprocal networks S21=3D3D3DS12 regardless =
of=20
>>=3D
>>   =20
>>
>the
>
> =20
>
>>sequence of the smaller building blocks within the network, BUT,=20
>>the=3D20 eye diagram shows us voltage transfer ratio between source =
and=20
>>load,=3D20 and it is NOT S21; the voltage transfer ratio DOES depend =
on=20
>>the=3D20 sequence of contributing blocks.  This happens also with =
linear

>>source=3D20 and load and even if the source and load are perfectly=20
>>matched.=3D3D20
>>
>>The only case when the voltage transfer ratio remains the same in=20
>>spite
>>   =20
>>
>
> =20
>
>>of moving building blocks around, if/when reflections at the=20
>>boundaries
>>   =20
>>
>
> =20
>
>>of the particular building block are negligible.
>>
>>Thanks
>>
>>Istvan Novak
>>SUN Microsystems
>>
>>
>>
>>Chris Cheng wrote:
>>=3D20
>>
>>   =20
>>
>>>I think the point is tuning S22 for the non-ideal load. S12 or S21 =
=3D
>>>     =20
>>>
>=3D3D3D
>
> =20
>
>>>=3D3D
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D20
>>
>>   =20
>>
>>>remains being equal (symmetric).
>>>
>>>-----Original Message-----
>>>From: lifeatthesharpend [mailto:lifeatthesharpend@xxxxxxxxx]
>>>Sent: Friday, September 28, 2007 2:44 PM
>>>To: signalintegrity@xxxxxxxxxxx; ron@xxxxxxxxxxx; Chris Cheng
>>>Cc: si-list@xxxxxxxxxxxxx
>>>Subject: Re: [SI-LIST] Re: AC Coupled Signals
>>>
>>>
>>>What is the total loss at each point? (reciever, driver side of cap,=20
>>>=3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D3D3D
>>
>>=3D20
>>
>>   =20
>>
>>>receiver side of cap) in dB=3D3D3D20
>>>
>>>It seems to stand to reason that if you lose xxx mV when the signal=20
>>>is
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D20
>>
>>   =20
>>
>>>=3D3D3D stronger at the driver and some smaller portion of that =
when=3D20

>>>the=3D3D20 caps =3D3D3D are placed closer to the recever is normal =
since=20
>>>=3D
>>>     =20
>>>
>loss=3D20
> =20
>
>>>has=3D3D20 ocurred in the =3D3D3D media. The total signal loss of the =
=3D
>>>     =20
>>>
>system=3D20
> =20
>
>>>could=3D3D20 still be the same.=3D3D3D20
>>>
>>>Leonard.=3D3D3D20
>>>
>>>
>>>
>>>----- Original Message ----=3D3D3D20
>>>From: Stephen Zinck <signalintegrity@xxxxxxxxxxx>=3D3D3D20
>>>To: ron@xxxxxxxxxxx; Chris.Cheng@xxxxxxxx=3D3D3D20
>>>Cc: si-list@xxxxxxxxxxxxx=3D3D3D20
>>>Sent: Friday, September 28, 2007 12:19:22 PM=3D3D3D20
>>>Subject: [SI-LIST] Re: AC Coupled Signals=3D3D3D20
>>>
>>>
>>>Hello SI-LISTers,=3D3D3D20
>>>
>>>I thought for my part in this discussion, I should do some=20
>>>due=3D3D20=3D20 diligence =3D3D3D on=3D3D3D20 this AC coupling =
capacitor=20
>>>placement =3D
>>>     =20
>>>
>location=3D3D20
>
> =20
>
>>>question.=3D3D3D20
>>>
>>>Scott McMorrow, Steve Weir and I had some off-line discussions that =
=3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D3D3D=3D3D20
>>=3D20
>>
>>   =20
>>
>>>tended=3D3D3D20 to suggest my position dependency results may have =
=3D
>>>     =20
>>>
>been=3D3D20
>
> =20
>
>>>caused by local=3D3D3D20 resonances from other impedance=20
>>>discontinuities =3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>in=3D3D20
>>=3D20
>>
>>   =20
>>
>>>the system I was=3D3D3D20 simulating. Based on this, I set out to =3D
>>>     =20
>>>
>develop=3D20
> =20
>
>>>=3D3D
>>>  =3D20
>>>
>>>     =20
>>>
>>a=3D3D20
>>=3D20
>>
>>   =20
>>
>>>simulation model that =3D3D3D had=3D3D3D20 a minimum of =
discontinuities =3D
>>>     =20
>>>
>(no=3D3D20
>
> =20
>
>>>backplane vias/connectors/trace, etc.). =3D3D3D
>>>
>>>
>>>I used:=3D3D3D20
>>>
>>>- Spice models of non-linear 3.125Gbit/s silicon (driver =
and=3D3D20=3D20=20
>>>receiver)=3D3D3D20
>>>- S-parameter based package models for both driver and =3D
>>>     =20
>>>
>receiver.=3D3D3D20
> =20
>
>>>- A 0.01uF capacitor and its associated parasitics (via, trace, pad,=20
>>>=3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D3D3D
>>
>>=3D20
>>
>>   =20
>>
>>>mount,=3D3D3D20 component).=3D3D3D20
>>>- 2D lossy W-Element transmission line (with di-electric and=20
>>>skin=3D3D20 =3D
>>>     =20
>>>
>
> =20
>
>>>effect =3D3D3D
>>>
>>>losses included).=3D3D3D20
>>>
>>>I made the capacitor model such that I could "slide" it up and=20
>>>down=3D20 a=3D3D20
>>>15 =3D3D3D
>>>
>>>inch trace between the driver and receiver. I iteratively =3D
>>>     =20
>>>
>simulated=3D3D20
>
> =20
>
>>>for =3D3D3D the=3D3D3D20 following length combinations:=3D3D3D20
>>>
>>>- 500 mil trace from driver to AC coupling capacitor with 14500 mil =
=3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D3D3D=3D3D20
>>=3D20
>>
>>   =20
>>
>>>trace to=3D3D3D20 receiver.=3D3D3D20
>>>- 5000 mil trace from driver to AC coupling capacitor with 10000 mil=20
>>>=3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D3D3D
>>
>>=3D20
>>
>>   =20
>>
>>>trace=3D3D3D20 to receiver.=3D3D3D20
>>>- 10000 mil trace from driver to AC coupling capacitor with 5000 mil=20
>>>=3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D3D3D
>>
>>=3D20
>>
>>   =20
>>
>>>trace=3D3D3D20 to receiver.=3D3D3D20
>>>- 14500 mil trace from driver to AC coupling capacitor with 500 mil =
=3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D3D3D=3D3D20
>>=3D20
>>
>>   =20
>>
>>>trace to=3D3D3D20 receiver.=3D3D3D20
>>>
>>>The results show around 125 mV (differential) difference=20
>>>between=3D3D20=3D20 the=3D3D3D20 capacitor at the source versus the=20
>>>capacitor at the=3D3D20=3D20 destination, with =3D3D3D the=3D3D3D20 =
benefit=20
>>>going to the capacitor=3D20 placed=3D3D20 closest to the receiver. =
125 mV

>>>is =3D3D3D a=3D3D3D20 lot to =3D
>>>     =20
>>>
>give=3D20
> =20
>
>>>away...=3D3D3D20
>>>
>>>I am not going to pretend to understand the physics behind=20
>>>these=3D3D20=3D20 results =3D3D3D but=3D3D3D20 I thought it worth =
while to at

>>>least show the =3D
>>>     =20
>>>
>=3D3D
> =20
>
>>>  =3D20
>>>
>>>     =20
>>>
>>basis=3D3D20
>>=3D20
>>
>>   =20
>>
>>>for my statements.=3D3D3D20
>>>
>>>I would be happy to evolve the simulation environment if someone=3D20 =

>>>has=3D3D20 a=3D3D3D20 suggestion...=3D3D3D20
>>>
>>>I have put together a document that I can post to an ftp site or=20
>>>email
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D20
>>
>>   =20
>>
>>>=3D3D3D if=3D3D3D20 anyone would like a copy...=3D3D3D20
>>>
>>>Kind regards,=3D3D3D20
>>>Steve=3D3D3D20
>>>
>>>Stephen P. Zinck=3D3D3D20
>>>Interconnect Engineering Inc.=3D3D3D20
>>>P.O. Box 577=3D3D3D20
>>>South Berwick, ME 03908=3D3D3D20
>>>Phone - (207) 384-8280=3D3D3D20
>>>Email - szinck@xxxxxxxxxxxxxxxxxxxxxxxxxxx=3D3D3D20
>>>Web - www.interconnectengineering.com=3D3D3D20
>>>
>>>
>>>
>>>
>>>  =3D20
>>>
>>>     =20
>>>
>>=3D20
>>
>>   =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:
>http://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:    =20
>               http://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
>
> =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:
http://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:     
                http://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