From: Mark Randol (email@example.com)
Date: Mon Nov 01 1999 - 13:09:48 PST
Bill Dempsey wrote:
> Mark Randol wrote:
> > 2nd, most ceramic 0603's and 0402's can have SRFs up around
> > 3-4GHz. Mounted parasitic inductances are another matter left
> > for the user :) The values top out at about 33pF for 1.5GHz, but
> > they are usable. Of course, I'm using them in a different
> > application than most on the list. I'm predominately concerned
> > with analog circuits, i.e. "discrete" frequency and relatively
> > low amplitude swings), so small capacitances are generally
> > sufficient. With digital circuits, I can understand where a
> > single small capacitance may not supply sufficient
> > "instantaneous" current. I don't know how much it costs to buy
> > and mount one of these, but from what I've been told by people
> > that should, 20 cents would be quite high unless you are using
> > porcelains or another special device.
> When you say "most 0603/402's can have SRF around 3-4GHz" you can NOT be referring to the typical ceramic X7R type decoupling caps used by "most" on this list. 0.1uF ceramics (0805) SRF around 30 MHz (depends...) and .001uFs
> ~150-200 MHz~. Add a long squirrley trace to this and watch the capacitor's effectiveness drop.... Todd Hubing published a good paper in the IEEE a few years back on the effectiveness of decoupling caps vs. board capacitance.
> It would behoove all to review this before making leaps of faith on decoupling strategies.
Sure I do Bill. What I said is that the value that's usable is
much smaller than what you might need, about 33pF. Also, note I
mentioned this small value might not be enough locally stored
charge to be useful in a digital application. Also^2 note I
mentioned in passing that HOW you place them is very important.
On chip decoupling (as someone else already mentioned) is better
if you can do it because you don't have to deal with the
inductance of the interconnects (wire bonds, packages, bumps,
etc). The draw backs and limitations are that it takes up
expensive die real estate PLUS they aren't very large values.
Faith in God? OK. Everything else, measure. (AKA In God we
trust, all others bring data.)
-- Mark Randol, RF Measurements Engineer Motorola SPS, Inc. M/S EL381 2100 E. Elliot Road Tempe, AZ 85284 (480)413-8052 Voice (480)413-3455 FAX M.Randol@motorola.com
**** To unsubscribe from si-list: send e-mail to firstname.lastname@example.org. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****
This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 11:39:29 PST