[SI-LIST] Re: Current Return Vias

Thought I'd throw in my 2 cents...

From my experience, there are 4 issues (at least) we are trying to
address with ground stitching vias, and they are very similar to the
issues involved in providing enough nearby ground pins on connectors and
packages. =20

The first, maintaining impedance, is relatively straightforward and
reproducible in 3-D simulations (though not easily - you must consider
the effect of your boundary carefully, and it's only applicable to the
specific stackup/geometries modeled).  To my simplistic thinking, it's
mostly related to the distance between the signal and the nearest ground
via.  A ground via nearby reduces the loop inductance and simultaneously
increases capacitance between the signal and ground, reducing the
impedance.  The tricky part, in my opinion, is that this may or may not
be a good thing.  If you are going through a board with a THICK core
(4-layer, 62mil), such that the via tends to appear inductive, relative
to the ciruit's nominal impedance (usually about 50 ohms), the nearby
ground via can be a good thing.  If you are going through a board with
LOTS of plances very close together such that the via appears
capacitive, the further reduction in impedance due to the nearby ground
via can make a situation worse.  This is true for both single-ended and
differential vias, though differential vias tend to have a
lower-than-optimal impedance, in my experience.  Only 3-D modeling
and/or measurement of test boards with your exact stackup/dimensions can
say what the preferred choice is to maintain proper impedance.  For a
133MHz bus, I doubt this will be a substantial factor (though I can
envision a signal via in the middle of a LARGE area without any ground
vias possibly being affected).

The second, reducing crosstalk due to several signals being forced to
use the same inductive ground path, is less straightforward to model
and/or measure (the model/measurement of the signal path must comprehend
the impact on the return path of nearby aggressors).  I haven't seen any
extensive work on this published for vias, though there should be quite
a bit available on the similar problem with connectors. Most future
high-speed designs go to differential signaling, which reduces the
problem (as it does for connectors, another plus for diff. signaling).
As others have alluded to, this could be troublesome for your design.
I'd look carefully to ensure several signals aren't forced to share the
same inductive return path.

The third issue that folks have raised is related to the first - trying
to maintain a nice TEM (or quasi-TEM) wave in the face of the ugly
reality of PCB construction, with its multiple planes, layers, and
necessary via physical design limitations.  I haven't seen any practical
way of doing this - I think we're going to have to absorb some ugliness
due to vias for a while.  Again, we hit this in connectors too (and the
interfaces to the connector), though we try to minimize the effects with
careful modeling and design.

A fourth, more subtle issue for differential signals, is trying to
maintain the same odd and even mode impedances as the transmission lines
leading into and out of the via (I.E., coupling of the vias is same as
that for the traces).  The position of the corresponding ground via(s)
plays a part here.  Fortunately, I don't think we'll have to worry about
this for a while - connectors don't pay much attention to it (that I
know of), and I haven't seen it highlighted as a problem for a design
solution space...

Disclaimer:
The content of this message is my personal opinion only and although I
am an employee of Intel, the statements I make here in no way represent
Intel's position on the issue, nor am I authorized to speak on behalf of
Intel on this matter.

Jeff Loyer
------------------------------------------------------------------
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 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:     
                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
  

Other related posts: