RE: [SI-LIST] : BGA Pkg Characterization

About this list Date view Thread view Subject view Author view

From: Larry Miller (ldmiller@rhapsodynetworks.com)
Date: Mon May 21 2001 - 06:49:13 PDT


Nope. The response characteristics depend upon how the routing from chip to
ball is done, and that would be a monstrous problem for a mathematical
analysis. It would also vary with manufacturer, unless they were all using a
particular package scheme.

The data I have seen is measured and curve-fitted with equivalent R-L-C
multi-section filters. There are trace-to-trace C's and mutual inductances
(transformers) and the resulting networks are definitely non-planar! Quite a
chore.

Some of these testing facilities that have responded here can give you
response sweeps (VNA runs) or TDR plots. I do not know if they offer the
services to turn these into SPICE models, for example. If you are working
below 300 MHz you probably wouldn't care. But 500 MHz to 2+ GHz is
definitely into the region where BGA parasitics ring loud and long.

Larry Miller

-----Original Message-----
From: Neeraj Pendse [mailto:cnepsc@ptg.nsc.com]
Sent: Friday, May 18, 2001 11:43 AM
To: Larry Miller
Cc: SI_LIST (E-mail)
Subject: RE: [SI-LIST] : BGA Pkg Characterization

Larry and anyone,

Have you found a way to predict resonance in BGA packages based on the
frequency, package size and substrate stack-up? And do you have any
advice on when I should be concerned about dying because of package
resonance?

Regards,

Neeraj.

National Semiconductor Corp.
http://www.national.com NYSE: NSM

>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 5/18/01, 6:22:48 AM, Larry Miller <ldmiller@rhapsodynetworks.com>
wrote regarding RE: [SI-LIST] : BGA Pkg Characterization:

> You really do need to do this if you are contemplating usage up above
300
> MHz. Many packages have simply awful resonances and crosstalk, and it
varies
> hugely with where you are on the package. I have found 4:1 variations
in
> lowest resonance frequencies in a single package due to the variations
in
> the way the signals get from the chip bonding pads to the balls. Also,
you
> need to test groups of signals. Various sets can resonate in pairs,
etc.

> Good luck!

> Larry Miller

> -----Original Message-----
> From: Wesley Wu [mailto:wesley@redswitch.com]
> Sent: Tuesday, May 15, 2001 8:23 PM
> To: SI_LIST (E-mail)
> Subject: [SI-LIST] : BGA Pkg Characterization

> Hi All,

> Has anybody done characterization on existing BGA packages recently? I
need
> to
> know if there is anyone who can do this for me and what is the
estimated
> cost. The
> required measurements include characteristic impedance along traces,
delay
> and
> gain on both directions of traces across the frequency band of 500M -
2GHz,
> ground
> bounce, cross talking, and ringing. I would like to obtain a data
range for
> all
> these to see if some particular BGA packages are suitable for my
> applications.
> Ideally I would like to add some equivalent to bond wire and IC input
> capacitance
> during test. Your help is greatly appreciated.

> Wesley Wu
> RedSwitch, Inc.
> (408)719-4748

> **** To unsubscribe from si-list or si-list-digest: send e-mail to
> majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
> si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu
> ****

> **** To unsubscribe from si-list or si-list-digest: send e-mail to
> majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
> si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu
> ****

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Jun 21 2001 - 10:12:01 PDT