[SI-LIST] Re: 3D EM -accuracy
- From: Cristian Gozzi <cristian.gozzi@xxxxxxxxx>
- To: bala89si@xxxxxxxxx
- Date: Tue, 19 Mar 2013 10:25:56 +0100
Hi folks
let me add something more about this topic.
You have to distinguish between 3D Field solver and which kind of Numerical
method is used to solve Maxwell´s Equations:
talking about 3D solver you have different kind of numerical methods and
approximation:
A) 3D Quasi-Static (BEM, DC FEM etc.)
B) 3D Full-Wave (FEM, FDTD etc.)
A) 3D Quasi-Static (like Ansys Q3D or Sentinel-NPE, Cadence ExtractIM and
others) approximates the field time variant to zero and extract E and H
field separately --> this will means a lumped RLCG extraction of your
interconnection
This lumped model does not take into account for any delay nor any skin
depth or radiation effect and usually it´s accurate enough if your
transition (VIA in your case) is small enough compared to wavelength (<<
lambda/6) or more conservative << lambda/10
this means that your field propagation inside the transition is almost time
invariant (quasi-static) so a lumped representation, without the HF
effects, is good
B) 3D Full-Wave (Like Ansys HFSS or Sentinel-PSI, CST Microwave Office
etc.) use the rigorous Maxwell´s equation to solve the transition effect
and the model is valid from DC up to the highest frequency of your freq.
sweep analysis
these tools take into account the delay effect, radiation and skin-depth
and they output S-parameter model (distributed model)
Usually it´s important to use these toosl when the transition electrical
dimension is more larger than lambda/6
I loaded in goolge/drive a short introduction to these topics
https://docs.google.com/file/d/0B9f_MeFAFByLVU4zYVMzX0hQQkE/edit?usp
=sharing
I hope this help
Regards
Cris
2013/3/19 Balaji G <bala89si@xxxxxxxxx>
> Balamanikandan,
> I believe Gert answered this question already. Its not that the 3D
> modeling improves accuracy at high speeds, 3D modeling MUST be done at high
> speeds. It is not that only recommended, It should be followed. A more
> accurate result depends on good modeling and that depends on taking many
> variables, boundaries in to account. In your intuition you can just guess
> that 3D modeling considers more. Hence at high speeds one may not trust the
> traditional rules of thumb & model, , say considering pi model for via.
>
> Regards,
> Balaji
>
>
> On Tue, Mar 19, 2013 at 11:35 AM, Balamanikandan K <kbmanick@xxxxxxxxx
> >wrote:
>
> > Hi,
> > I know very well that 3D EM modelling is recommended for via and
> > connectors etc if we go for a high speed design.
> >
> > I would like to know how the accuracy is improved if we use 3D EM
> when
> > the frequency is in GHZ.
> >
> > Please explain the accuracy with frequency with respect to 3D EM.
> >
> >
> > --
> > Best Regards,
> > Balamanikandan.K
> >
> >
> > ------------------------------------------------------------------
> > 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
> >
> >
> >
>
>
> ------------------------------------------------------------------
> 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
>
>
>
------------------------------------------------------------------
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
Other related posts:
