A good way to think about this, especially on a larger size board, is to imagine
a transmission line resonator. Assume for the moment that the shielding
enclosure and the board ground are tied together at one point. In this case, the
board will go resonant when the length of the board ground measured from the tie
point becomes 1/4 of a wavelenth. This is not that long of a distance at even a
low order harmonics of 100 MHz. All kinds of nasty things happen with such a
resonance, the worst being radiated/conducted emissions, though circuit
operation of sensitive nodes can be affected as well.
Don Borowski
Schweitzer Engineering Labs
Pullman, WA
David Heald <dhe...@tellium.com> on 10/11/2002 12:53:27 PM
Please respond to David Heald <dhe...@tellium.com>
To: "'emc-p...@majordomo.ieee.org'" <emc-p...@majordomo.ieee.org>
cc: (bcc: Don Borowski/SEL)
Subject: grounding schemes & EMI
All,
I'm trying to convince a few people here that completely separating the
digital and chassis grounding on our product is not always the best way to
go. Unfortunately, a lot of the people I'm dealing with are ex Bellcore
engineers who worked a lot with isolated grounds and are convinced that
isolated grounds are the only way to go. Now we're dealing with optical
interfaces and speeds well in excess of 100MHz, so I really want to see the
grounds tied together as much as possible.
While I know that combining the digital and chassis grounds is for the most
part better once you get above a few hundred MHz, putting together concrete
arguments is proving to be a bit elusive. I luckily have some high level
backing that will let me push my views, but I am one person up against a
team of industry vets.
If anyone has been in this boat before and won, could you share some of the
tactics or arguments that you used? I know this issue has been discussed in
the past, but a fresh discussion of the relative benefits of isolating the D
and Cgnds would probably be beneficial to the group as well. See below for
my views on the issue.
Thanks!!!!
Dave
My views for telecom equipment with a backplane and plug in circuit packs
(and a good tight chassis around it all):
(Note that Analog grounds are outside of the scope of this statement - I'm
focusing on Digital grounds and Chassis ground)
The benefits of separating Dgnd and Cgnd have to do with defining your
signal impedances and SI in general. When you place this system inside a
Cgnd "balloon", all should be well but maybe there is some extra noise due
to RF being trapped within the balloon.
However, if the Cgnd and Dgnd are tied together throughout the system, the
effect should be similar to "heat shrinking" your conductive chassis Cgnd
ballon onto your Dgnd. The single ended signal return currents should still
follow their original paths and things should essentially remain unchanged.
I could see some possibility (I'll avoid use of the word potential here :o)
) for RF currents on the circuit pack card grounds due to RF fields
contained within the faraday cage, but I think these could be mitigated by
clever bonding of the grounds on circuit packs. I think that isolating the
faceplate from the Dgnd on the circuit packs but stitching the bottom edge
(faceplate to backplane) Cgnd ESD guard band to Dgnd could alleviate stray
currents on the cards and keep them relatively clean - all while still
maintaining the bonding of the Cgnd and Dgnd on a system level. The idea
(as my brain developed it) is to keep the stray currents at the periphery of
the card by limiting the through connections on the circuit packs and
forcing stray currents to flow near the edge of the card. The backplane
should for the most part have Dgnd and Cgnd be one and the same.
Does this raise any red flags for anyone? I'm expecting at least a few, but
this is the best scheme that I can come up with right now.
-------------------------------------------
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/
To cancel your subscription, send mail to:
majord...@ieee.org
with the single line:
unsubscribe emc-pstc
For help, send mail to the list administrators:
Ron Pickard: emc-p...@hypercom.com
Dave Heald: davehe...@attbi.com
For policy questions, send mail to:
Richard Nute: ri...@ieee.org
Jim Bacher: j.bac...@ieee.org
All emc-pstc postings are archived and searchable on the web at:
http://ieeepstc.mindcruiser.com/
Click on "browse" and then "emc-pstc mailing list"
This e-mail may contain SEL confidential information. The opinions expressed
are not necessarily those of SEL. Any unauthorized disclosure, distribution or
other use is prohibited. If you received this e-mail in error, please notify
the sender, permanently delete it, and destroy any printout. Thank you.
-------------------------------------------
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/
To cancel your subscription, send mail to:
majord...@ieee.org
with the single line:
unsubscribe emc-pstc
For help, send mail to the list administrators:
Ron Pickard: emc-p...@hypercom.com
Dave Heald: davehe...@attbi.com
For policy questions, send mail to:
Richard Nute: ri...@ieee.org
Jim Bacher: j.bac...@ieee.org
All emc-pstc postings are archived and searchable on the web at:
http://ieeepstc.mindcruiser.com/
Click on "browse" and then "emc-pstc mailing list"
