[SI-LIST] Re: Book - (Chassis - signal ground)
- From: Chris Padilla <cpad@xxxxxxxxx>
- To: si-list@xxxxxxxxxxxxx
- Date: Fri, 22 Jun 2001 09:15:13 -0700
Okay, Larry called so I must chime in here...
I agree with chassis connections using the old lambda/20 rule but this
isn't necessarily true for PCB stitiching as far as reducing EMI goes.
By PCB stitching, I am assuming that (please correct my interpretation if I
am wrong) you mean ground vias moving around the periphery of the PCB to
help reach a Faraday cage effect, correct? If so, read on....
I have seen 3D simulations of multi-layer PCBs showing that energy bouncing
back (i.e. reflecting off the periphery via, sign reversal) can cause
problems as well by coupling the energy into other vias that help send it
out the top/bottom of the board through traces and ICs. So which is
worse: energy leaking out the edges of PCBs or energy leaking out the
tops/bottoms of PCBs? I haven't a clue and the simulation didn't provide
answers to that question. This is more food for thought. I have to admit
that seeing the energy bouncing around between planes on a PCB was
disquieting but perhaps better there than leaking out the edge?
Perhaps some discussion (and a new thread topic) will shed some light on
ways to absorb energy at the periphery of PCBs? I have heard of resistors
being used at boards edges to try and accomplish this.
What do I do in practice? I still ask for ground via periphery stitching
on all my boards (mostly high-density, hi-speed ethernet).
----->Chris
Chris Padilla
EMC Engineer
Cisco Systems
>If your board stitching and chassis connections are placed at a distance of
>lambda/20 you reduce EMI, not enhance it. The only time you can get away
>with single point grounding is at very slow edge rates. What you will do
>with single point is increase your conducted emissions due to copper losses.
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at: //www.freelists.org/archives/si-list
Old list archives are viewable at: http://www.qsl.net/wb6tpu
Other related posts:
