Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re:RE: 8b/10b and EMI packaging

Dear Wayne and Edward

Here are some comments from another reflector lurker

Although it is easy to say EMI is "black magic", a good understanding of
effective high-frequency grounding and shielding practices including the chassis
as well as the board layout from the beginning of the design will avoid the
unneccessary headaches.  At frequencies above 1.0 GHz one has to have a back
ground in disciplines in microwave and antenna theory to stay ahead of the game.
 In Gigabit Ethernet a laser diode driver may have a rise times on the order of
200 picoseconds (20%-80%) to generate the required risetime with a reasonable
margin.  In the EMI test chamber there can be measureable harmonics up to 5.0
GHz and real power (to easily fail FCC class A) to 3.0 GHz or so.  

Here is an example of microwave EMI. A inexpensive chassis design is made with
front panel slots cut to accept inexpensive optical connectors (like gigabit
ethernet dual SC connectors).  The front panel to mother board and transceiver
ground path distance is on the order of 6" (15 cm).  Since the transceiver laser
diode is probably in close proximity this design - especially the front panel -
is a good receipe for a slot antenna or a slot antenna array.  A ground 6" away
is no ground at all since that is a full wavelength at 2.0 GHz where Gigabit
ethernet data transceiver rise times can easily generate strong harmonics.  One
must know how to find sources of radiation and identify unwanted antenna
structures, and then to silence and shield the radiation sources and "de-Q" the
antenna structures for cost-effective EMI solutions at microwave frequencies. 
This know-how and expertise has to be added to the necessary grounding and
shielding practices to take care of EMI radiation at less than 1000 MHz.  Of
course a small form factor - cutting slot dimensions in half  - is a good thing
since it moves the "slot antenna" resonant frequencies above the radiating
source, killing the EMI antenna.  Double the transceiver speed and now one is
back where one started from with radiating source frequences at double speed now
matching the half dimension small form factor slot.  It sounds like you (Edward
and Wayne) have had a little exposure to these matters and do not believe its
all "black magic".


Robert Friedman
Molex Fiber Optics
tel: 630 512 8756
fax: 630 810 8981 

____________________Reply Separator____________________
Subject:    RE: 8b/10b and EMI packaging
Author: "Edward Chang" <edward.chang@xxxxxxxxxxxxxxxx>
Date:       03/15/2000 5:14 PM


You are right, no question about it.  As long as it is cost-effective, all
parts including component package, board layout, circuit design, cable
dressing ... should use the good common practice to avoid the unnecessary
EMI headache.


Edward S. Chang
NetWorth Technologies, Inc.
Tel: (610)292-2870
Fax: (610)292-2872

-----Original Message-----
From: Wayne Belshaw [mailto:wab20@xxxxxxxxxx]
Sent: Wednesday, March 15, 2000 4:28 PM
To: Edward Chang
Cc: stds-802-3-hssg@xxxxxxxx
Subject: RE: 8b/10b and EMI packaging

Comment from a reflector-lurker:

Regarding EMI design, I should point out that the most
cost-effective system designs take EMI characteristics
into the basic board designs to such a degree that EMI
is reduced at the source as much as possible.  This
not only improves the stability and manufacture of the
designs, but also allows for cheaper packaging materials
for the enclosures (which can really affect today's
competitive edge on appearances and cost - use of
plastics and the like).

Just my two-cents worth!

Best to you,
Wayne Belshaw
Amdahl Corporation

At 01:40 PM 3/15/2000 -0500, you wrote:
>Tom and ALL:
>Usually the board is inside a cabinet.  The EMI test is based on a system,
>but not a component or a board.  The enclosure is usually well designed to
>shield all radiations both in and out.  The only possible leak is the
>cables.  In an optical system, a cable is not a concern at all.
>Although, if the transceivers are not well designed for EMI concerns, and
>excessive radiation parts may cause some headache; however, in general, the
>8B/10B code has been used for many years, and I have not heard anyone had
>EMI problem.  I never had experienced any EMI problem at all in my
>The key is to design a enclosure being well shielded from radiation.
>Edward S. Chang
>NetWorth Technologies, Inc.
>Tel: (610)292-2870
>Fax: (610)292-2872
>-----Original Message-----
>From: owner-stds-802-3-hssg@xxxxxxxx
>[mailto:owner-stds-802-3-hssg@xxxxxxxx]On Behalf Of Tom Truman
>Sent: Wednesday, March 15, 2000 11:13 AM
>To: Ed Grivna
>Cc: stds-802-3-hssg@xxxxxxxx
>Subject: Re: 8b/10b and EMI
>Thanks for the response.
>If 8b/10b were to be scrambled, then it would appear
>to me that all it is providing at the XAUI interface is packet delineation
>and some error monitoring capability. I imagine that each lane would need
>a separate scrambler/descrambler, initialized to different states so that
>the transitions across the lanes are uncorrelated. Synchronizing these
>and deskewing the lanes would require some thought -- it isn't difficult,
>but it isn't as straightforward as the "alignment column" proposed for
>At that point, the 25% overhead of the 8b/10b scheme
>seems to be a staggering price to pay for delineation and
>error monitoring -- why not start with scrambling, at a lower baud rate,
>make the overall design problems simpler?
>Best regards,
>Tom Truman
>Lucent Technologies
>Ed Grivna wrote:
>> Hi Tom,
>> a good source of data onthis can be found in the IEEE 1394b development
>> archives.  Since 1394b maskes use of 8B/10B encoding, and they spend
>> a lot of effort on EMI reduction, there should be a significant number
>> of papers/presentations available on the subject.
>> As a background on what I remember, it is definately possible to
>> create "hot spots" in the radiated emission spectrum if you keep
>> repeating a short sequence of characters.  This occurs in Fibre Channel
>> systems with their 4-character Idle sequence.  To get around this,
>> 1394b added a level of scrambling to both the source data cahracters
>> AND to the command characters used, prior to sending them through the
>> 8B/10B encoder.  By doing this they were able to achieve some dramatic
>> (sorry, I can't remember the number of dB) reduction in radiated
>> emissions.
>> The 8B/10B code, when sending random data, has a fairly wide emissions
>> spectrum (which is what you want), but if you sit on the same
>> character or small group of characters, you can see the discrete
>> spectral peaks quite clearly.
>> Regards,
>> Ed Grivna
>> Cypress Semiconductor
>> >
>> > I would like to raise the issue of EM emissions with 8b/10b
>> > vs. scrambling (spectrum comparisions can be found in the SLP
>> > or in Joel Goergen's presentation). My impression is that component
>> > are leaving this problem for the system integrators to solve.
>> >
>> > Has anyone looked into the issue of EMI/EMC within the context of
>> > cost and time-to-market?
>> >
>> > While system integration and implementation are
>> > beyond the scope of the standard, it would be grossly negligent to
>> > these issues, as ultimately the goal is to get product out the door
>> > satisfies FCC/ETSI regulations.
>> >
>> >
>> > Tom