[SI-LIST] Re: AW: Re: Reference layers for high speed diff p

From: Sen Velmurugan <sen.velmurugan@xxxxxxxxxx>
To: si-list@xxxxxxxxxxxxx
Date: Thu, 24 Jan 2013 11:45:13 -0800
I like to put some  more 3D mindset on this in the 2D email. The 
localization of issue needs to be more fine grain. So I want to put 
forward my thinking and understanding.

The questions should not be "Quite GND plane",  The question should be 
"Quite signal Ref plane SEGMENT" and low impedance return  path for your 
signal path. When there is a high magnitude return current in  the 
adjacent segment, your signal's reference segment experience crosstalk 
degradation. So a field solver WHICH see such situation can help you to 
know your margin.

I don't know which filed solver simulates the adjacent segment current 
flow. For example if there is DC-DC converter GND as the adjacent GND 
segment for your reference segment ?
But for a group of parallel traces in a bus, it can do an SSO simulation.
But how about the circuit on the other side of the GND plane, how much 
return current they are pumping ? Then, Is it OK to stitch two GND 
planes in that vicinity for multi multi layer PCB.
So, the reference segment for signal  is affected by adjacent segments  
circuit on top and bottom of the PCB. That's why we try to use more and 
more GND layers in a PCB.

Also, GND plane is a universal reference plane but DEPENDING on the I/O 
driver construction it's power plane can be used as a 
reference(DDR,SSTL). But power plane is generated by a power source, 
many circuits contend for its energy, so it is not at steady potential 
(ripple, droop), so it is not a steady reference as a whole and it has 
segment effect as explained above.  Hence SI margin is more with GND
even when the PI of the power plane is great. As such does PI tool 
models the power source the way it will behave ?

So your reference segment's absolute E&M wave magnitude and its phase 
w.r.t signal matters. So
there is no "Reference point" available or needed to measure noise on 
the GND segment.
~Sen

------------------------------------------------------------------
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 forum  is accessible at:
               http://tech.groups.yahoo.com/group/si-list

List archives are viewable at:     
                //www.freelists.org/archives/si-list
 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu
[SI-LIST] Re: AW: Re: Reference layers for high speed diff p

Other related posts:
