[SI-LIST] Re: Power delivery model with real GND ports ?
- From: Siming Pan <pansiming86@xxxxxxxxx>
- To: Aubrey Sparkman <asparky@xxxxxxxxxxxxxxxxxxxxxxxx>
- Date: Tue, 23 Nov 2010 09:30:27 -0800
Hi Motti,
I agree with Aubrey. Although the electrical potentials may be different
at different GND locations in the real world, a lumped GND node can be
always given mathematically. For example, in PCB with GND vias, we can
always assume the voltages across those GND vias are zeros. By adding these
conditions, all the inductances associated with these GND vias may be
lumped into one single element. Although this mathematic manipulation is not
physical, sometime it may help to build the model.
Regards,
Siming
2010/11/23 Aubrey Sparkman <asparky@xxxxxxxxxxxxxxxxxxxxxxxx>
> Steve,
>
> Let me ask you to reconsider based on your own comments. IF the customer
> is
> asking for a more accurate package model from Intel, they would probably
> want all the node 0 connections to be in the PCB with Vdd and Vss
> connections somewhat like a signal, because, as you said below, using a
> node
> 0 connection in the package would short out real power distribution issues.
>
> There was some work in the mid 90s with simulation and measurement
> techniques where a temporary ground was used for peripheral leaded parts
> such that all pins were treated as signals which allowed a full matrix
> without the use of node 0. Look for stuff related to JEDEC JC-15. Also,
> there was some work and a paper by Bryan Young on the math to transform the
> inductance matrix from one with a temporary ground to one where the matrix
> was reduced by identifying specific pins as ground. But at the time,
> "single lump" models were sufficient.
>
> I think the technique for using a temporary ground for the package is still
> valid, but I haven't tested recently. Certainly the transform math gets
> way
> more complex when the phase difference across the package is no longer very
> small at the bandwidth frequency.
>
>
> Aubrey Sparkman
> 512-461-6165
> asparky@xxxxxxxxxxxxxxxxxxxxxxxx
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
> On
> Behalf Of steve weir
> Sent: Tuesday, November 23, 2010 8:10 AM
> To: Haller, Motti
> Cc: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: Power delivery model with real GND ports ?
>
> I think the problem is with your customer's concept of ground. Your PDN is
> going to exhibit voltage drops from wherever you define your reference Vdd
> and Vss points. If you attach node 0 connections in more than one place
> you
> will in effect short out parts of your model and it will not report correct
> results.
>
> If you just want voltage measurements referenced to SPICE node 0 for
> convenience, insert E elements for those measurements.
>
> Steve
>
>
> Haller, Motti wrote:
> > Hi
> > Does anyone know how to create a power delivery model without losing the
> GND ports?
> >
> > My end user want to have the gnd/vss (in & out) ports not embedded as ref
> but as real ports.
> >
> > (can be touchstone / spice or RLGC).
> >
> > If I use this gnd source/sync port as ref. How does the gnd resistance
> and
> inductance reflects into the VCC* ports/nets.
> >
> >
> > Thanks
> > Motti
> >
> > ---------------------------------------------------------------------
> > Intel Israel (74) Limited
> >
> > This e-mail and any attachments may contain confidential material for
> > the sole use of the intended recipient(s). Any review or distribution
> > by others is strictly prohibited. If you are not the intended
> > recipient, please contact the sender and delete all copies.
> >
> > ------------------------------------------------------------------
> > 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
> >
> > Old (prior to June 6, 2001) list archives are viewable at:
> > http://www.qsl.net/wb6tpu
> >
> >
> >
> >
>
>
> --
> Steve Weir
> IPBLOX, LLC
> 150 N. Center St. #211
> Reno, NV 89501
> www.ipblox.com
>
> (775) 299-4236 Business
> (866) 675-4630 Toll-free
> (707) 780-1951 Fax
>
>
> ------------------------------------------------------------------
> 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
>
> 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 technical documents are available at:
> http://www.si-list.net
>
> 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
>
>
>
--
Siming Pan
------------------------------------------------------------------
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
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
