Re: [SI-LIST] : Surface Mount Cap Lead Inductance

Ray Anderson (raymonda@radium.eng.sun.com)
Tue, 11 Aug 1998 12:12:19 -0700 (PDT)

Mark-

Way back when I worked at Contel ASC developing 14 GHz freq. multipliers
for satcom applications, one of our purchasing people discovered that
you could save a couple of tenths of a cent per part by buying unmarked
ATC caps in lieu of the laser marked parts. The mfg. people had a hard
time getting the parts mounted consistantly in the correct orientation
as the parts were nearly cubical and it wasn't readily obvious what
was vertical and what was horizontal orientation. The end result was
that some of the GAsfet matching networks failed to tune correctly
about half the time. Yes, electrode orientation can make a difference,
but fortunately, in most SI applications this hasn't proved to be a
concern (yet). I think this is in the same category as observing
differences in narrowband filter resonances at microwave frequencies
depending on how your printed traces are positioned relative to the
weave of the fiberglass fibers in glass/teflon board material. At
sufficiently high frequencies things that are insignificant at lower
frequencies become very important.

-Ray
Sun Microystems

>
> This isn't lead inductance per say, but since we've drifted a
> bit to what can contribute to resonances:
>
> 3 - The orientation of the cap as mounted to the PCB can make
> a difference too. ATC (American Technical Ceramics -
> www.atc-cap.com) uses a mulitlayer construction, and suggests
> the plates be oriented perpendicular to the PCB surface. This
> reduces component to board capacitance <?>. I've made resonance
> problems disappear and VCO's resonate in band by just rotating
> the caps from "flat" to "vertical". (ref ATC Bulletin #201,
> Recommended Attachment Techniques for Multilayer Capacitors)
>
> ATC is the only manufacturer I know of that has orientation
> markings, so maybe others construct theirs differently?
> --
> ---------------------------------------------------------------
> Mark Randol, RF Measurements Engineer | Motorola SPS, Inc.
> (602)413-8052 Voice | M/S EL379
> (602)413-4150 FAX | 2100 E. Elliot Road
> ryvw50@email.sps.mot.com | Tempe, AZ 85284
> ---------------------------------------------------------------