[SI-LIST] Re: FW: Re: Cap discontinuity

No need to be that humble, Jeff...I'll just ignore the part about =
impedance being 0 at all frequencies.  The rest of your message was =
accurate and helpful as far as my understanding goes; the parasitics of =
the cap ARE the primary effects causing a disconituity on the signal =
path, particularly the parallel capacitance of the pads, unless the =
inductance of the body is enough to cancel them out.  Everybody types =
faster than they think sometimes.  I find posting to the SI list is the =
#1 way to remind myself of this.

Jeremy

-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Loyer, Jeff W
Sent: Friday, May 17, 2002 7:19 AM
To: 'Signal Integrity Mailing List'
Subject: [SI-LIST] FW: Re: Cap discontinuity



Please ignore much of my response.  As someone kindly pointed out, even
ideal caps have impedance.  Physics is infuriatingly humbling.

Jeff Loyer


If you could put an "ideal" capacitor in-line, there would be no
discontinuity.  The "ideal" capacitor would have no physical length,
resistance, or inductance.  Its impedance at all AC frequencies would be =
0.

If you put this "ideal" capacitor on pads which are wider than the
transmission line, the pads decrease your characteristic impedance, =
causing
an impedance discontinuity.  The best approach would be to put the =
"ideal"
cap. on the trace directly, without affecting the characteristics of the
trace.

With a non-ideal capacitor, you are now going to add finite length,
resistance, and inductance.  These will add to the impedance =
discontinuity
produced by the addition of the cap.  They will be there regardless of =
how
you mount the non-ideal cap to the trace.  Adding further parasitics by
widening your trace for the attachment is probably a bad thing.  An
exception might occur if the parasitics of the cap and the parasitics of =
the
pads could be considered a lumped element.  In that case, if you KNOW =
that
your non-ideal cap. produces a higher impedance than your trace's
characteristic impedance (it appears more inductive), you might =
compensate
for this with a more capicitive trace.  2 caveats: at 10G, I doubt your =
cap
and pad parasitics can be lumped; and without careful characterization =
of
your cap., you're as likely to make things worse as better. =20

Jeff Loyer

-----Original Message-----
From: mittalr@xxxxxxxxxx [mailto:mittalr@xxxxxxxxxx]
Sent: Thursday, May 16, 2002 10:41 AM
To: jian@xxxxxxxxxx
Cc: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Cap discontinuity


 Thanks for your comments, Jian.

However, I still don't know the answer - does the width of the pad =
versus
the trace matter. I know the cap will cause discontinuity, as you =
mentioned
below. Can we safely assume that the discontinuity from the pad is much =
less
than the the cap. I can't find a simple simulation to do to verify.

Thanks




--- On Thu 05/16, Jian X. Zheng  wrote:
> Hi, Rohit:
>=20
> I would like to make some comments on this subject.
>=20
> From electromagnetics, any discontinuities in a uniform transmission =
line
> will create the reflection in the transmitting power. The reflection
> normally is higher and higher when the size of the discontinuities is
> getting larger and larger compared to wavelength.
>             -----
> -----------/     \--------------
> -----------\_____/--------------
> incident   A     B
>=20
> At low frequency, the difference between the pad size and the TLN may =
not
> cause much reflection because the reflection from A and the reflection
> from
> B are out of phase and they will cancel each other when the distance
> between
> the A and B is small. However, the reflection will become higher when =
the
> distance between A and B is no longer very small compared to =
wavelength.
> Thanks!
>=20
> Best regards,
>=20
> =
-----------------------------------------------------------------------
> Jian-X. Zheng, Ph.D
> Zeland Software, Inc., 48890 Milmont Drive, 105D, Fremont, CA 94538,
> U.S.A.
> Tel: 510-623-7162, Fax: 510-623-7135, Web: http://www.zeland.com
> ---------------------------------------------------------------------
>=20
> > -----Original Message-----
> > From: si-list-bounce@xxxxxxxxxxxxx
> > [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of mittalr@xxxxxxxxxx
> > Sent: Thursday, May 16, 2002 9:00 AM
> > To: si-list@xxxxxxxxxxxxx
> > Subject: [SI-LIST] Cap discontinuity
> >
> >
> >
> > Lot of books/app notes mention that if you have high speed lines,
> > say 10G then you should make the pad size of the ac decoupling
> > cap the same width as the trace.
> >
> > Can anyone give the reason why? It seems if you put a cap in a
> > 10G line, there will be reflection anyway because the Z of the
> > cap is not the same as the Z of the line. Then how much effect
> > does matching the width of the pad to the line have anyway
> >
> > Thanks
> > Rohit
> >
> > ------------------------------------------------
> > Join Excite! - http://www.excite.com
> > The most personalized portal on the Web!
> >
> >
> > ------------------------------------------------------------------
> > 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:
> > http://www.freelists.org/webpage/si-list
> >
> > For help:
> > si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
> >
> > List archives are viewable at:
> >             http://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

------------------------------------------------
Join Excite! - http://www.excite.com
The most personalized portal on the Web!


------------------------------------------------------------------
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:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:    =20
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages=20
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:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:    =20
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages=20
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:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:    =20
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages=20
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:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                http://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: