[SI-LIST] Re: ESD is a low frequency event -really??
- From: MikonCons@xxxxxxx
- To: si-list@xxxxxxxxxxxxx
- Date: Mon, 15 Mar 2004 15:59:18 EST
In a message dated 3/12/2004 4:59:34 PM Pacific Standard Time,
weirsp@xxxxxxxxxx writes:
Mike, if you would be so kind as to indulge me here, I would like further
details on the sort of situation that justified BC for ESD control. Can
you cite ballpark numbers where absent BC, the plane impedance was too high
and a destructive event was only correctable by pulling the impedance down
with BC?
********
Steve: I'll attempt to be detailed enough in my response to hopefully put
this subject to bed. Some of my comments are not directly solely to you, but to
the less initiated on on SI List.....
On more than a dozen boards that were experiencing both soft and hard
failures, using BC was ONE of the correcting factors that eliminated all ESD
susceptibility. As I noted in my earlier comments on this thread, I use
multiple
techniques in consort to correct deficiencies in a given board. When I am
called in
to correct an EXISTING problem, the client is usually in deep schedule
trouble, is losing their clients, and/or is losing market share. They demand
(and
need) an immediate and absolute fix for their problems with an immediate
turnaround. The typical project I am called in to FIX has already spent four to
six
weeks trying to correct deficiencies with in-house staff, and used from one to
three other (lower priced) consultants (at $85-$125/hour) who have also failed
to correct the problems (sometimes over several months). My rates are 3-4
times higher, but I only spend 4-32 hours to correct the problems (100% success
over 19 years, >98% client repeat rate). I'm usually then retained to critique
4-10 other tasks/projects based on my successful resolution of their problems.
I have guided the design of hundreds of NEW boards with the goal of total
compliance first time--every time in both functional performance and all
regulatory compliance requirements with a success rate of >98% on the first
shot and
100% with an iteration on the errant few.
Specifically for ESD, I have generally coupled the low impedance, sacrificial
nature, of multiple-layer chassis ground rings around the board edges with BC
(preferably two sandwiches) as an effective fix. Note that I do NOT insist on
either of these techniques if the operating environment does not portend the
need. If there are other layout deficiencies that I find, I of course
recommend additional corrective action. I always define a sequencial order of
corrective actions with an assigned priority of expected beneficial impact,
generally
supported with modeling and analysis to bound performance. Invariably, I have
achieved correction for the original problems and also reduced radiated
emissions by 6 dB to >30 dB (the latter for boards with poor initial layouts).
The technical justification for using BC for reduced ESD susceptibility is
multifaceted. First, it yields a low PDS impedance over a very broadband range
which helps to control the IC electrical operating environment for common-mode
voltage (CMV) injection above ~40 MHz. If you have ever done work with RF
tuned tanks, you will be aware that a high C:L ratio for a given resonance will
yield lower Q circuits. This is a major (and unheralded) benefit of BC. The
thinner the dielectric between the planes, the greater the impact of the
dielectric losses. The low-Q effect is to lower the response to physically
dictated
board resonances. And yes, of course, multiple bulk capacitors and selected
small
ceramics are still required to handle the lower frequency range, and they
will also impact where those resonances occur. Nevertheless, the bottom line is
that the low-Q resonances offered by BC sandwiches help prevent unexpected and
abberant EMI behavior.
Now, from the view of the IC, the power plane, the ground plane, AND the
signal reference planes (since they should be ground or power) are all tightly
coupled with BC usage so that any injected CMV is strongly attenuated above ~40
MHz. THIS IS A VERY GOOD THING!!!! The IC tends to see ONLY normal-mode and/or
differential-mode signals as intended, even in the presence of externally
injected transients (ESD events) AND other impinging radiation that is then
observed by the IC as CMV!!! For perfect PDS and signal decoupling (close, but
no
cigar), the IC would not see any perturbation.
Think about it.
Mike
Michael L. Conn
Owner/Principal Consultant
Mikon Consulting
Cell: 408-821-9843
*** Serving Your Needs with Technical Excellence ***
------------------------------------------------------------------
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:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl
List technical documents are available at:
http://www.si-list.org
List archives are viewable at:
//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:
