[SI-LIST] Re: I have a question related package parasitic extraction
- From: "Michael Khusid" <mkhusid@xxxxxxxxxx>
- To: <cnepsc@xxxxxxxxxxx>
- Date: Tue, 9 Dec 2003 15:25:36 -0500
Neeraj,
Ansoft Spicelink has two modules. The 2D module, which Jayaprakash =
refers
to, is the one that solves for per-unit parameters calculates loop
inductances.
The 3D modules, which KD refers to, is typically used for 3D problems =
like
wirebonds. The latter module does indeed solve for partial inductances.
KD,=20
If you have very long trace with uniform cross-section, then indeed
capacitance/inductance/resistance are linear with length. That's how
original telegrapher equations were derived.
If there is a discontinuity between two signals, e.g. a trace and =
wirebond,
than mutual interaction (inductance and capacitance) will produce
non-additive results. That is, the properly done addition also needs to
include mutual interaction.
> As for L, there should be and there is difference. Small difference =
also
> expected for C.
Neeraj, you are ignoring the effect of mutual capacitance which is =
non-zero
in this case.
Best regards,
Michael
----------------------------------------------------------------
Michael Khusid
Ansoft Corporation
SI/HF Application Engineer
=20
25 Burlington Mall Road, 5th floor
Burlington, MA 01803-4100
=20
Tel 781-229-8900 Ext. 134=20
Fax 781-229-8624
---------------------http://www.ansoft.com ---------------------
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx =
[mailto:si-list-bounce@xxxxxxxxxxxxx]
> On Behalf Of Neeraj Pendse
> Sent: Tuesday, December 09, 2003 1:56 PM
> To: edpc108@xxxxxxxxx
> Cc: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: I have a question related package parasitic
> extraction
>=20
> Jayprakash:
>=20
> The definition of inductance that the transmission line equations use =
is
> different from that used by Ansoft SPICELINK. The former is Loop
> inductance, which scales with length but the later is Partial =
inductance
> which does not because of self coupling.
>=20
> KD:
>=20
> I can think of two reasons for this difference:
> #1: source and sink assignments in the two models, especially the
> contact of the lead with the wire in the "separate" models
> #2: Skin effect: Did you solve AT the freq of interest or did you =
solve
> at the default 100 MHz and "reduced" to the freq you want? In the =
later
> case, one has to be careful because the "reduction" result is =
generated
> by simple formula and is not valid for all frequencies in my opinion.
>=20
> As for L, there should be and there is difference. Small difference =
also
> expected for C.
>=20
> Also, there will be some difference due to the fact that the mesh on
> each object in the three cases will be different because of the
> differences in geometry. I think this error in a simple case can be
> reduced by tightening the convergence criterion to let's say 0.2% or =
so.
>=20
> Hope that helps.
>=20
> - Neeraj.
>=20
> Jayaprakash wrote:
> > Hello Patrick,
> >
> > Then the same argument should hold good even for
> > Transmission line model described using telegrapher's
> > equation?
> >
> > By telegrapher's equation, Transmission line has
> > uniform RLGC per unit length. The output voltage is
> > given by
> >
> > Vo =3D exp(-gamma*Line_length)*vin (for a perfectly
> > matched system)
> >
> > Is this description of Transmission line not really
> > scalable? (Going by the explanation given below, I
> > expect over projection of loss for longer lines than
> > actually it is)
> >
> >
> > Thanks and regards,
> > Jayaprakash.
> >
> >
> > --- Patrick Zilaro <pzilaro@xxxxxxxxxxxx> wrote: >
> > KD,
> >
> >>There is nothing wrong with the commercial solver.
> >>Even if you were to
> >>simulate a planar structure such as a trace, you
> >>would see that the "parts"
> >>do not add up to the "whole". For example, if you
> >>simulated a 5mm trace and
> >>a 10mm trace using the exact same stack-up, etc. you
> >>would find the
> >>following:
> >>
> >>1) capacitance -> 2*5mm trace should be pretty equal
> >>to 10mm trace
> >>2) (ac) resistance -> 2*5mm trace should be less
> >>than the 10mm trace
> >>3) inductance -> 2*5mm trace should be even more
> >>less than the 10mm trace
> >>
> >>This is because each segment of the structure
> >>interacts with the rest of the
> >>segments. So be careful, when you break up a
> >>problem, especially since you
> >>are introducing error on the optimistic side (where
> >>you usually would prefer
> >>to be on the pessimistic side).
> >>
> >>Regards,
> >>
> >>Patrick
> >>
> >>-----Original Message-----
> >>From: si-list-bounce@xxxxxxxxxxxxx
> >>[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of KD
> >>KIM
> >>Sent: Monday, December 08, 2003 3:59 PM
> >>To: si-list@xxxxxxxxxxxxx
> >>Subject: [SI-LIST] I have a question related package
> >>parasitic
> >>extraction.
> >>
> >>
> >>Dear all,
> >>I have a question and be happy to write to you.
> >>
> >>I have simulation the structure composed of only one
> >>bonding wire and one
> >>lead,
> >>using FEM commercial solver, Ansoft Spicelink like
> >>below for extraction
> >>parasitic values,
> >>R, L, and C.
> >>
> >>I have simulated for below three cases.
> >>
> >>1. the structure contained only one bonding wire
> >>2. the structure contained only one lead
> >>3. the structure composed of one wire and one lead
> >>
> >>I think the value for case 3 equal to the values
> >>case 1 added case 2.
> >>But the result was not. The difference was about 16%
> >>for the Resistance.
> >>Of course, other parasitic values also have some
> >>differences considerably.
> >>Is this difference resulted from the defect of the
> >>FEM commercial solver?
> >>I want to know that what do you think about it?
> >>Please reply to my e-mail.
> >
> >
> >
> > =3D=3D=3D=3D=3D
> > Balachandran Jayaprakash
> > Celestijnenlaan,
> > 3/61, 3001, Heverlee,
> > Leuven, Belgium.
> > Mobile: +32-472-630120
> >
> > =
________________________________________________________________________
> > BT Yahoo! Broadband - Save =A380 when you order online today. Hurry! =
Offer
> ends 21st December 2003. The way the internet was meant to be.
> http://uk.rd.yahoo.com/evt=3D21064/*http://btyahoo.yahoo.co.uk
> > ------------------------------------------------------------------
> > 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.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
> >
> >
> >
> >
> >
>=20
>=20
> --
> National Semiconductor Corporation
> 2900 Semiconductor Drive, M/S 19-100
> Santa Clara, California 95051
> http://www.national.com/, NYSE: NSM
>=20
>=20
>=20
>=20
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>=20
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>=20
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>=20
> List technical documents are available at:
> http://www.si-list.org
>=20
> 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
>=20
------------------------------------------------------------------
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.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:
