Re: [SI-LIST] : What's your favorite Screwy SI Concept?

About this list Date view Thread view Subject view Author view

From: Michael Vrbanac (vrbanacm@swbell.net)
Date: Tue Jan 11 2000 - 21:09:13 PST


To Michael Chan,

I am not sure what you mean by "drawn up". If its a "physical description"
you are looking for, perhaps the following will help:

Starting with two adjacent power/ground distribution planes, the entire power
plane perimeter is designed to be smaller than the ground plane by a factor of
20 times the distance between the two planes when measured at any point
along the ground plane perimeter edge.

Hope that is what you meant!

To the forum,

It looks as if we have found what the 20H rule is and who originated it
and who wrote something about it. That's good progress but we need
more. It is obvious that its effectiveness is disputed, so let's continue
to explore the possible reasons why the 20H rule might exist for as long
as it has. Without making any claims one way or other, I will offer some
"thought experiments" to hopefully help us along.

I used to do this with my team some years ago when I was the technical
leader/manager of an very well equipped electromagnetic laboratory at
a former employer. It was a very effective technique and we solved some
pretty interesting and complex problems with it (and the lab).

We first have to assume that there was a beneficial intent by the application
of the rule and it was thought to have been effective in its original use.
Otherwise, the whole thing would be an intentional sham and we should
forget about it altogether. Since there would be no benefit of throwing it
out without adequately researching any possible ways it could be of benefit,
let's assume that it had some perceived well intended benefit from the
application of it.

Then by logical deduction, we could conclude that a few things might happen
as a result of this 20H rule (from a general point of view) just from a
physical description of it and applying some basic knowledge of physics to
see if there are any plausible avenues of understanding the 20H rule's usefulness.

(Note I am not claiming that any of these "logical deductions" are actually
true. I am trying to help us find the reason why it exists.)

1. Since it involves a reduction of physical size in an electric conductor with
a consistent surrounding dielectric(s) (by definition), it "generally" means that the
intrinsic resonant frequency of one of the smaller plane will be increased. It is not
always true but is generally true (I am thinking of complex shapes here being the
possible exception). (A question: What might be the intended effect of making
the two planes have two different primary resonant frequencies?)

Therefore: there could have been issues with regards to planar resonances
which the creation of this rule may have originally been attempting to address.
There also may be a "high" Q effect thought to have been a factor in the
original problem and by changing one of the elements of the problem, perhaps
this effect could have been mitigated.

2. Since the planes are of differing voltage potentials (by definition), there is
an opportunity to observe a change in fringing E-field patterns at the plane
edges.

Therefore: there could have been issues with regards to E-field impingement
on some nearby circuit of some sort that the rule was originally attempting
to address.

3. Since it is not difficult to conceive that power and ground plane circulating
currents can be generated by active devices near the edges of the planes, the
actual levels at a distance of 20H would be logically thought to be significantly
smaller, so there might be an opportunity to observe a downward change in
inductive coupling levels. If the use of the rule would also include the
restriction of signal routing to within the power plane perimeter dimensions,
then this would also indeed seem to be true simply by virtue of the added
distance.

Therefore: there could have been issues with low level inductive coupling
in the original situation that could have prompted the use of this rule.

I'll stop with three instances. Perhaps there may be other observations
out there which could help us understand the existence of the 20H rule.
Please offer them. Perhaps this will "stimulate" thinking on how to
tackle this "beast" from another vantage point.

Notice that I have not made an attempt to "prove" anything. I am after
"understanding" which may later lead to a "proof" if there is one.

BTW, maybe I should give this process a fancy name, write a paper, and
claim I created it but alas, I'll bet some intelligent person thought of it
before! <grin> I might call it something impressive like "The Scientific
Method" or some such. <grin>

Enjoy!

Michael E. Vrbanac

"Chan, Michael" wrote:

> The question still unanswered. How is the 20H rule being drawn up?
> I don't care whether it is a good or bad rule. I just hope to know
> how it is being drawn up. Hope this won't offend the other audiences
> in this forum.
> I do agree with Doug's explanation of using rule of thumb. I use other
> people's rule of thumbs as well but I will try to verify it before it use
> it.
>
> -----Original Message-----
> From: Grasso, Charles (Chaz) [mailto:GrassC@LOUISVILLE.STORTEK.COM]
> Sent: Tuesday, January 11, 2000 6:03 PM
> To: 'si-list@silab.eng.sun.com'
> Cc: 'Doug Smith'
> Subject: RE: [SI-LIST] : What's your favorite Screwy SI Concept?
>
> Doug, an excellent answer - very eloquently put.
>
> -----Original Message-----
> From: Doug Smith [mailto:dsmith@corp.auspex.com]
> Sent: Tuesday, January 11, 2000 3:46 PM
> To: si-list@silab.eng.sun.com
> Subject: Re: [SI-LIST] : What's your favorite Screwy SI Concept?
>
> Hi All,
>
> Buryed within the multiple responses to the 20H rule was the following
> text:
>
> > In the acknowledgements amongst other *Celebrities* is the name of DOUG
> SMITH,
> > for providing Technical review of the material. If he is the same as the
> one
> > on this list we may get some clarity over this.
>
> At the time Mark was writing his first book, he did ask me to review it
> and I started doing so. However, reviewing a book is a lot of work. At
> the time, my work life was very intensive and did not permit me the time
> to finish a review for Mark in the time he needed. I was able to provide
> some input to Mark and felt honored that he asked me for it. It is a
> sign of a good engineer/scientist to take into account technical input
> from others, and Mark did that to his credit.
>
> I build up experiments, time permitting, to test out theories and see
> what kind of measured results I can get. The results of the latest board
> layout experiment have been submitted to the IEEE EMC Symposium and we
> (more than one author) hope to present the material next summer at the
> Symposium. I have not done such an experiment yet for the 20H rule and
> so have no data to present. With all the discussion, is anyone planning
> such an experiment in the near future?
>
> A word on rules of thumb... We all use rules of thumb every day in our
> work and daily lives. These come from many sources and we cannot
> possibly check them all out. For instance, Henry Ott's book is full of
> rules of thumb and summaries of principles such as the equation for
> radiation from a hole. Henry's book would have to be VERY long to
> provide the proof of all of them. But, they are reasonable and come from
> a knowledgeable source so we use them. Sometimes we misapply a principle
> here or there but for the most part such an approach works (or we would
> all be looking for work OR get very little sleep trying to check out
> everything we read). If a rule of thumb makes sense given one's own
> knowledge and experience and it comes from a reliable source, use it. It
> is always good to generate our own data or proof on these matters, but
> that is not always practical. Isn't this characteristic, that of
> building on the work of others involving some trust, what distinguishes
> human beings from animals, the ability to pass on information without
> each generation having to continually regenerate it for themselves?
>
> What really makes this work was said by Carl Sagen:
>
> ~"A good engineer or scientist must be willing to drop their most
> cherished theories in the face of credible evidence to the contrary."
> Note: ~".." means approximate quote. Maybe if I took more Ginko Biloba I
> would remember the exact quote. :-)
>
> Doug
>
> --
> ----------------------------------------------------------
> ___ _ Doug Smith
> \ / ) Manager EMC Development & Test
> ========= Auspex Systems
> _ / \ / \ _ 2300 Central Expressway
> / /\ \ ] / /\ \ Santa Clara, CA 95050-2516
> | q-----( ) | o | Phone/FAX: 408-566-2157/2020
> \ _ / ] \ _ / Email: dsmith@auspex.com
> ----------------------------------------------------------
>
> **** To unsubscribe from si-list: send e-mail to
> majordomo@silab.eng.sun.com. 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
> ****
>
> **** To unsubscribe from si-list: send e-mail to
> majordomo@silab.eng.sun.com. 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
> ****
>
> **** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. 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
> ****

**** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. 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
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:34:34 PDT