From: Greim, Michael ([email protected])
Date: Mon Jun 05 2000 - 13:25:01 PDT
I think that you may find some difference of opinion on
this front, but from my perspective an analytically selected
group of caps can help minimize the peaks and valleys
of PDS impedance that can appear when only one or two
different (but widely spaced values) are utilized. One of the
issues that really doesn't get alot of attention is the effective
inductance of the capacitor and its attachment to the PWB.
While ESR will determine the depth of the notch the ESL
will determine how quickly the impedance falls off. Impro-
perly attached (high L) low ESR caps will often realize these
high impedance notches at frequencies between the resonance
frequencies of your decoupling. This will manifest itself as
PDS failure when the current draw frequency falls at these notch
frequencies.(often code dependant). That is why it is important to know
the impedance target of your PDS and know that your decoupling
scheme stays within budget.
I recommend performing a frequency sweep of your decoupling
scheme to see if it meets your goals. While one or two values
may be easier from a mfg perspective. Your impedance budget
targets may only be met with a mix of three or four cap values.
If your two cap values satisfy your impedance budget then you
are all done.
Remember, your mileage may vary.....
And all this science they don't understand
It's just my job six days a week.....
The time is gone, The email's over, thought I'd
something more to say.........
Michael C. Greim Consulting Engineer
Mercury Computer Systems, Inc email: [email protected]
199 Riverneck Road V: 978-256-0052/x1607
Chelmsford, MA 01824-2820 F: 978-256-4778
> -----Original Message-----
> From: Jay Marano [SMTP:[email protected]]
> Sent: Monday, June 05, 2000 3:00 PM
> To: [email protected]
> Subject: [SI-LIST] : RE: bypass caps
> I'm looking for any feedback from those who have
> read the white paper on the Ultracad web site
> regarding ESR and Bypass self resonant behavior.
> One of the things it demonstrates is that for a
> given number of capacitors, better system impedance
> results are obtained from using more cap values,
> with moderate but not super low ESR, spread over a
> range than with a smaller set of cap values, with
> very low ESR, at well chosen self-resonant frequecies.
> I understand the theory behind his analysis, but it
> raises an interesting question: In the frequency range
> between the bulk capacitors and the knee frequency of your
> system, is it better to purposely use wider tolerance caps
> and use a smaller number of values, or buy more tighter tolerence
> values and rely on the nominal values to smooth out the system
> impedance? Any feedback on this analysis would be appreciated.
> Jay Marano
> Sr. Hardware Design Engineer
> Maitland, FL 32751
> [email protected] <mailto:[email protected]>
> **** To unsubscribe from si-list or si-list-digest: send e-mail to
> [email protected] 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
[email protected] 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
This archive was generated by hypermail 2b29 : Wed Nov 22 2000 - 10:50:31 PST