Ken, Lee,
Yes, what I meant was that in mixed-signal boards and on some boards
where naturally very noisy circuits (like high-current DC-DC converters)
have to coexist with sensitive circuits, like SerDes, PLLs, etc,
splitting the ground plane between the domains can improve the isolation
and thus improves performance, as long as we make sure that we dont
create SI and/or EMI problems by crossing the splits with signals that
otherwise need a solid reference. For instance, when I open up my USB
powered 100Ms/s oscilloscope, its PCB clearly has moats, cuts and slots
on the planes (including the GND or reference planes) between the ADC
input section and the high-swing arbitrary waveform generator DAC circuit.
Regards,
Istvan Novak
Oracle
On 1/17/2017 6:32 PM, Ken Wyatt wrote:
Argh, my apologies Istvan. I indeed misread your description, as pointed out
by two of our colleagues, Vinu and Bert.
Yes, I agree that there are cases where planes may be split for various
reasons - mainly isolation, etc. We just need to be sure not to cross that
split with clock or other high-speed traces.
Crawling back into the woodwork… :-)
Cheers, Ken
_______________________
I'm here to help you succeed! Feel free to call or email with any questions
related to EMC or EMI troubleshooting - at no obligation. I'm always happy to
help!
Kenneth Wyatt
Wyatt Technical Services LLC
56 Aspen Dr.
Woodland Park, CO 80863
Phone: (719) 310-5418
Email Me! <mailto:ken@xxxxxxxxxxxxxxxx> | Web Site
<http://www.emc-seminars.com/> | Blog <http://design-4-emc.com/>
The EMC Blog (EDN) <http://www.edn.com/blog/The-EMC-Blog>
Subscribe to Newsletter
<http://www.emc-seminars.com/Newsletter/Newsletter.html>
Connect with me on LinkedIn <http://www.linkedin.com/in/kennethwyatt>
On Jan 17, 2017, at 1:05 PM, Ken Wyatt <ken@xxxxxxxxxxxxxxxx> wrote:
Hi Istvan,
I would have to agree there can be many trade-offs in board designs -
especially in very dense designs. You mentioned some cases where splits were
OK and I’d be interested in any additional specific conditions where
splitting the return (or reference) plane is OK from a SI or EMI point of
view, or can even make it better, as you point out.
In general, my experience has been in cases where high speed (>100 kHz, say)
clocks or other digital signals cross a gap in the return plane, common mode
harmonic currents are produced, which can create not only crosstalk to
victim traces, but worse, these currents can flow out the outside of I/O
cable shields, creating the potential for radiated emissions. I understand
there are situations where this can’t be helped and I do believe there are
cases where this doesn’t matter much.
I created a short video demonstration of this effect.
https://youtu.be/L44lTnQgv-o
Cheers, Ken
_______________________
I'm here to help you succeed! Feel free to call or email with any questions
related to EMC or EMI troubleshooting - at no obligation. I'm always happy
to help!
Kenneth Wyatt
Wyatt Technical Services LLC
56 Aspen Dr.
Woodland Park, CO 80863
Phone: (719) 310-5418
Email Me! <mailto:ken@xxxxxxxxxxxxxxxx> | Web Site
<http://www.emc-seminars.com/> | Blog <http://design-4-emc.com/>
The EMC Blog (EDN) <http://www.edn.com/blog/The-EMC-Blog>
Subscribe to Newsletter
<http://www.emc-seminars.com/Newsletter/Newsletter.html>
Connect with me on LinkedIn <http://www.linkedin.com/in/kennethwyatt>
On Jan 16, 2017, at 9:05 PM, Istvan Novak <istvan.novak@xxxxxxxxxxx> wrote:
Lee,
Nice article and I agree with most of what you say, though I still would
not use extreme phrases, like on page 6 you say "Splitting a ground
plane destroys its integrity as an ultralow impedance connection between
all of the components in a circuit and should never be done." I agree
that the reference (ground) plane is very important and we should make
sure it can serve its purpose properly. However, in complex and high
density designs there are several conflicting requirements and sometimes
making careful splits on the ground planes are not only OK, it will make
the product outright better. There are occasions, for instance, when we
need to equalize and optimize the DC drop on planes and in addition to
optimizing the placement of components, we can also help by splitting
the planes, power and/or reference planes. If signal traces are
naturally routed alongside the split without the need to cross it, the
signaling will be OK. There are real, working products with split
ground planes.
Regards,
Istvan Novak
Oracle
On 1/16/2017 6:53 PM, Lee Ritchey wrote:
A few weeks ago we had a prolonged exchange on what ground is. With------------------------------------------------------------------
Martin
Rowe, we did an article that appeared on the UBM site with this URL.
http://ubm.io/2ikEDat
I wrote another article that includes this as well as an expanded section
on
EMI and ground and Faraday Cages and how they contain EMI. That article is
now posted on the Speeding Edge web site, www.speedingedge.com
<http://www.speedingedge.com> for any who want to go beyond what the first
article covered.
Hope both of these help clear up some of the confusion around this topic.
See you all at DesignCon in Santa Clara at the end of the month.
By the way, I am planting flowers in my ground this month!
Lee Ritchey
Speeding Edge
P.O. Box 2194
Glen Ellen, CA
95442
707-568-3983
I just used the energy it takes
To get mad and wrote some blues.
Count Basie
Or:
Worry is like a rocking chair,
It keeps you busy but.
It doesn't get you anywhere.
------------------------------------------------------------------
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 forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
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 forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
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 forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
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 forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu