Re: [SI-LIST] : 20-H Rule for Power Planes

John Lockwood (johnl@juniper.net)
Thu, 27 May 1999 18:28:03 -0700

Hi Bill,

I'm responding to your comment about tricking the E/M waves. In the cases
where I have fixed EMI failures with this technique( refer to my previous
posting) all I really did was move the source further away from the trace,
air vent, or card guide to which it was coupling. We are talking about very
small spacings and near field here. In the near field the magnitude drops
as a squared or cubed function, so small increases in spacings can create
big improvements.

John Lockwood
Juniper Networks

At 06:44 PM 5/27/99 -0500, you wrote:
>I must be dumb....
>
>To understand this 20H problem better let me break down my
>understandings
>and see where I went wrong.
>
>EMI = Electro Magnetic Interference
>
>Hmmmm... Okay.
>
>I've got the electric field part down (i.e. the +++ plane with
>reference to the --- plane makes an E field).
>Kinda reminds me of a PP capacitor.
>
>Gotta have the magnetic part to get an EM wave. Of course if I'm
>just interested in E field disturbance then I guess I don't care. But
>I'm searching for EM disturbances....so I'll look elsewhere.
>
>The "M" (or H) portion as I recall is caused by a change in current.
>(Quick
>check to my brain recalls a name of Gauss so I'll use that to impress
>you
>all out there). Okay, granted. A noise source is present on one plane
>with reference to
>the other. This can cause a change in my E field.... What exactly is
>the
>strength of the H field in a PP capacitor with a time varying noise
>source.
>(I honestly can't say I know that). Current is flowing in the planes
>which
>generates what size H field in the presence of my E field?
>
>How do I "trick" my E/M waves (where's the M again?) from jumping off my
>board by cutting back the +++ plane? That one's got me fooled. I think
>Mr. Maxwell needs a seance session with me and I'll ask him.
>
>If anyone can explain this in basic terms I'd love to hear it. Or
>better
>yet if you've got some near field and far field measurements on boards
>with
>and without the 20H rule in effect I'd love to see the results.
>
>Help!
>
>Bill Dempsey
>DNA Enterprises, Inc.
>
>p.s. is Cisco hiring?
>
>**** 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 ****