[SI-LIST] Re: Analog and Digital grounds

  • From: "Dimiter Popoff" <dp@xxxxxxxxxxx>
  • To: "si-list" <si-list@xxxxxxxxxxxxx>
  • Date: Tue, 13 Sep 2005 02:10:47 +0300

Separate ground planes is something one would seldomly need, simply
because there typically are not that many signal lines between the digital
and the analog parts of a board and it is indeed easier to keep them
just physically apart. So this "new" rule is well suited for engineers who
are not sure what they do with the grounds etc., it will keep them
from messing things up, and will do a good job in most cases.
 Having said that, splitting the ground plane in addition to keeping the
digital and analog blocks apart can be useful. If crossing the splits is an 
issue
(fast repetitive signals), the signals must connect the two realms (A and D)
above the interconnecting ground plane part.
 Splitting the plane will be mostly useful against lower frequency
interferences due to high current values (think SSO and
some repetitive algorithm).

------------------------------------------------------
Dimiter Popoff               Transgalactic Instruments

http://www.tgi-sci.com
------------------------------------------------------




>  -------Original Message-------
>  From: Stuart Brorson <sdb@xxxxxxxxxx>
>  Subject: [SI-LIST] Re: Analog and Digital grounds
>  Sent: Sep 13 '05 01:00
>  
>  >
>  > When I was coming up (longer ago than I care to admit to) I was taught th=
>  > e conventional wisdom that Analog and Digital signals should have seperat=
>  > e ground traces (later ground planes) which connect at a single point.  I=
>  >  have gathered that this is as old fashioned as using a single 0.1 uF cap=
>  > acitor for each power pin.
>  > =20
>  > Will some kind people point me to some papers or other resources where I =
>  > can learn what the current thinking is (and the theory behind it).  The l=
>  > atter is essential because trying to get some people to abandon the "trie=
>  > d and true" methods for something more up to date takes a lot of convinci=
>  > ng. =20
>  
>  I first heard that multiple GND planes is a no-no at a National
>  Semiconductor analog seminar a couple of months ago.  The speakers
>  said that you are better off having one GND plane for analog and
>  digital, and just keep the analog and digital components as physically
>  separated as possible.  They said that the chances that you'd mess up
>  the AGND and DGND return currents -- with signal traces running over
>  ground cuts, and what not -- was larger than any noise improvement you
>  might get using separate planes.  This makes sense to me since any
>  analog/digital design I have been involved with has so many traces
>  running every which way that it's impossible to maintain separate AGND
>  and DGND planes underneath all nets without creating a gerrymandered
>  mess.  (And before I am attacked for using bad placement, please be
>  assured that I do try to place the components as logically as possible
>  before routing begins!)
>  
>  Here's a pointer to some references I pulled off the National website
>  talking about this:
>  
>  http://www.national.com/kbase/category/Data_Conversion.html#57
>  
>  You can search their site for more material.
>  
>  HTH,
>  
>  Stuart
>  ------------------------------------------------------------------
>  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: