[SI-LIST] FEXT reduction by very small spacings
- From: "Eric Bogatin" <eric@xxxxxxxxxxxxxxx>
- To: <hermann.ruckerbauer@xxxxxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Tue, 18 Aug 2009 07:03:05 -0500
Hermann- before you jump to too radical conclusions about the impact of very
tight coupling on FEXT, you might want to think about another field solver
tool.
Agilent's ADS has a built in 2D boundary element field solver, in the MIL
structures. This is a super tool to use to sweep through geometry features
and explore the impact on performance. However, it assumes the currents are
in sheets parallel to the infinite return plane. It does include the skin
depth and frequency dependent inductance from current redistribution, but
the current density distribution is an approximation.
When the traces are closer than their width, I would question how good this
approximation is. You might want to re-spin this simulation using a tool
that calculates the current distribution in the cross section, like
Agilent's Momentum, EMPro (the 2D port solution that uses an FEM approach),
or one of the other tools like CST, HFSS or Ansoft's SI2D
When considering tight or loose coupling in diff pairs, if loss is
important, you might consider loose coupling, as you suggest, but if
interconnect density is important and loss is not, nothing beats tight
coupling. This and other diff pair design topics is covered in my upcoming
free webinar,
NMA-830 Stack up Design for Differential Pair Design, Sept 16, 2009 1 pm
Eastern time
Sign up on line @ www.beTheSignal.com <http://www.bethesignal.com/>
See you in cyberspace
--eric
*******************************************************
Dr. Eric Bogatin, Signal Integrity Evangelist
Bogatin Enterprises
Setting the Standard for Signal Integrity Training
Next "No Myths Allowed" webinar:
"Stack-up Design for Differential Pairs"
Sept 16, 2009 1pm EDT
Next public classes: EPSI, ASID
Aug 11-14, 2009, Hillsboro, OR
<http://www.bethesignal.com/> www.BeTheSignal.com
Blog: <http://www.bethesignal.com/blog> www.beTheSignal.com/blog
26235 W 110th Terr
Olathe, KS 66061
e: <mailto:eric@xxxxxxxxxxxxxxxx> eric@xxxxxxxxxxxxxxx
v: 913-393-1305 cell: 913-424-4333 f: 913-393-0929
***********************************************
Msg: #2 in digest
Date: Mon, 17 Aug 2009 21:54:58 +0200
From: Hermann Ruckerbauer <hermann.ruckerbauer@xxxxxxxxxxxxx>
Subject: [SI-LIST] Re: FEXT reduction by very small spacings
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Hi Istvan,
thanks for the detailed information. Your paper is very interesting!
I'm still not sure if a design with 100um Spacing could not be as good
as a design with 300-400um spacing. Even with the manufacturing
tolerances the X-talk behavior might not get worse, and this would still
mean, that the space requirements are going down quite a bit. I guess in
the 90s manufacturing a 75 or 100um spacing was a bigger problem as it
is today, so it might be worth taking a new look to the effect ?!?
But now there are already 3 things that would need to be checked quite
carefully:
- NEXT
- Min AND Max. spacing for ALL lines
- Manufacturing tolerances
=3D=3D> Overall I agree, that this looks quite ugly and I'm not sure if a=
useful implementation is possible and worth the effort.
One more question comes into my mind: What does this effect do on
differential routing ?
Some years ago I would have routed a diff pair as tight coupled as
possible ... today I think loose coupling with a solid referencing is
better.
But I never considered, that this effect would be seen in the tight
coupled case.
But I guess here also the manufacturing tolerances make things
"unreliable", so a loose coupling is better, as this effect is not seen
there.
Regards
Hermann
EKH - EyeKnowHow
Hermann Ruckerbauer
www.eyeknowhow.de
hermann.ruckerbauer@xxxxxxxxxxxxx
Veilchenstrasse 1
94554 Moos
Tel.: +49 (0)9938 / 902 083
Mobile: +49 (0)176 / 787 787 77
Fax: +49 (0)721 / 151 258 230
------------------------------------------------------------------
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
Other related posts:
