[SI-LIST] Re: Is Impedance Enough for Describing the PDS?

  • From: steve weir <weirsp@xxxxxxxxxx>
  • To: "Abhijit Mahajan" <amahajan@xxxxxxxxxxxx>, si-list@xxxxxxxxxxxxx
  • Date: Fri, 09 Jul 2004 01:40:01 -0700

Abhijit, yes localization is an approach that can be very helpful in 
reducing or completely eliminating cavity resonance issues.  For a long 
time we have been able to get away with letting planes perform dual 
duty:  wide low impedance lumped conductors for power, and the return side 
of signal transmission lines.  But we are long past the point where signal 
spectra is way beyond board lambda / 4. for common plane size / Er 
combinations.  Continued dual-use under these circumstances is an 
invitation to EMC nightmares, and greatly complicates PDS realization.  But 
it does keep consultants like me busy.  Sigrity, and Ansoft sell tools to 
try and manage these problems.  The tools are very nice, but I think the 
tools address the wrong problem.

If we "simply" eliminate the cavities ( OK shrink them so that they are not 
problematic ), as you suggest then I think we are on the right 
track.  There is documentation out that clearly shows the economic benefits 
of islanding power.  In one recent paper, the author demonstrated that a 
board with islanded power and 20mil power / ground separation performed 
almost identically EMC-wise to the same size assembly using 2mil BC.  The 
tool and material vendors probably don't like hearing this.

Islanding power presents challenges with stack-up planning and routing.  We 
still have to:

1) Create a local low impedance on the island, and
2) Get the DC current to the island

I see the tool opportunities in supporting these efforts, as the physics 
and economics will really demand it.  On the low-impedance island technique 
end, I am a big fan of low inductance capacitors and optimized attachment 
techniques.  The really low inductance capacitors that can be fitted to a 
PWB today are:

IDC's, very expensive
X2Ys, more than cheapy 0603 MLCCs but a lot less than IDCs
NEC Proadlizer ( consult NEC on price )

If done correctly, islanding can save a bundle, and goes directly to one of 
Chris Cheng's pet mantra's that B/C is a solution looking for a 
problem.  We can avoid such problems and therefore the cost of B/C.

Steve.
At 01:06 PM 7/9/2004 +0530, Abhijit Mahajan wrote:
>Steve/Others.
>
>Would it be better then to reduce the area of power planes as much as
>possible(by
>localizing the power planes to only cover the pins that need it) and pack
>the decoupling there?
>islands of same voltage can be then connected with wide traces to the
>nearest bulk cap
>And finally brought together at the regulator.
>
>This can be vaguely be described as a "star topology" for power
>distribution.
>
>Does anyone think there is any value in this appproach?  Has it been
>discussed before?
>
>
>##I am assuming all signal routing is over a common ground plane.
>
>
>Thanks!
>
>Abhijit.
>
>
>-----Original Message-----
>From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
>Behalf Of steve weir
>Sent: Friday, July 09, 2004 6:04 AM
>To: zhang_kun@xxxxxxxxxx; si-list@xxxxxxxxxxxxx
>Subject: [SI-LIST] Re: Is Impedance Enough for Describing the PDS?
>
>
>Zhangkun,  the problem is that the PDS overlaps both signal integrity and
>EMC compliance.  If the system rules are done right, then the PDS
>decoupling problem will eventually reduce to an impedance profile.  But
>that profile is both spatial and frequency dependent.  If I shove all the
>decoupling capacitors into one corner of the board I get a very different
>result than if I distribute them.
>
>Steve.
>At 08:04 AM 7/9/2004 +0800, zhangkun 29902 wrote:
> >Dear all
> >
> >When talking about power delivery system with other engineer, there are
> >a
> >lot of element to describe the PDS, such as energy storage, decoupling,
> >bypassing, and so on. I think all these parameter could be described by
> >one element, impedance of PDS.
> >
> >Is my point right? Is impedance enough for describing the PDS?
> >
> >Best Regards
> >
> >Zhangkun
> >2004.7.9
> >
> >------------------------------------------------------------------
> >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 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:
> >                 //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
> >
>
>
>------------------------------------------------------------------
>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 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:
>                 //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
>


------------------------------------------------------------------
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 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:     
                //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: