Mitch:
The following is taken directly from Rick Hartleys "Control of EMI and
Crosstalk athe the PC Board Level", as well as "High Speed Printed Circuit
Design - A Practical Approach"
"Current Paths - Every circuit is a closed loop. When a device drives a
current down a wire to another device, after being received, a current of
equal value will be returned to the sending device, creating a closed loop.
In a PC Board environment, when a signal propagates down a trace to an IC,
the path of the initial signal is defined by the position of the signal
trace. If the return current is low frequency, (up to a few hundred KHz.),
the return current will follow the path of least resistance, usually the
shortest and/or widest path. Once the signal gets above a few hundred KHz,
but certainly by the low MHz range, the return signal will always attempt
to tightly couple both its electric field and its magnetic field with the
original signal. This means returning to the source as close to the path
of the original signal as possible. When a trace is routed directly over a
ground plane, the return current will flow in the plane directly under the
original signal path, back to the source. At MHz frequencies and higher,
the signal return will follow the path of least impedance, which usually
means the path of least inductance."
>Dave,
>
>I must disagree with this "concept", but do agree with hi-speed lines
>crossing splits, but only regarding impedance control.
>
>This "concept" of the "return path" is what I don't agree with. Yes, I
>know it is being "taught" now in certain circles, but I don't find any
>mention to it in Lee Ritchey's classes (Lee's what the industry considers
>the hi-speed guru, and disagrees with the "return path" theory), nor in
>James Blankenhorn's SMTPlus hi-speed classes, nor in Rick Hartley's
>hi-speed classes.
>
>I did attend a class(es) by Mary Sudgren recently (the Copper
>Connection), and she went into the return path theory again, and seemed
>like a no-issue (regarding the above mentioned "experts"), and took
>away time from some real issues (let's talk NOISE BUDGETS!).
>
>With that said, I believe the real issue regarding split planes is most
>recognizing proper power distribution and impedance issues.
>
>BTW, you might be seeing an article on the "return path" theory in the
>near future in PC Design magazine authored by Lee Ritchey, so keep
>your eyes open.
>
>Mitch Morey
>NSI Communications
>San Diego, CA
>www.nsicomm.com
>
>>>> David Baldwin <[log in to unmask]> 07/21/98 08:38am
>>>>
>The biggest problem is when high speed lines cross the split in the
>planes.
> The return currents follow directly under the traces, and if there is a
>split in the plane, the currents have to flow around the split causing
>emissions problems. There is no real problem splitting the plane if high
>speed lines do not cross that barrier.
>
>Dave.
>
>At 10:13 AM 7/21/98 -0700, you wrote:
>>At the expense of sounding ignorant, what are the problems caused by
>>splitting gnd planes? I have to do this constantly, mainly under
>>opto-isolators, to seperate the part of the board that is our product from
>>the outside world. I hope you can get some online answers so that
>everyone
>>has a chance to check the data. Thanks.
>>Richard
>>
>> ----------
>>From: dschaefer
>>To: DesignerCouncil
>>Subject: [DC] !!! SPLIT GROUND PLANES !!!
>>Date: Tuesday, July 21, 1998 8:25AM
>>
>>I'm looking for information regarding SPLIT GROUND PLANES.
>>
>>Experts on grounding, EMI, EMC are teaching that gaps in ground planes
>>create serious problems. Component manufacturers, however,
>continue to
>>recommend placing gaps under components (such as A/D's and D/A's).
>>
>>Can anyone:
>>
>> - point me in the direction of theoretical reference material
>> - relate their experiences / practices
>>
>>Thanks,
>> --
>>Dave Schaefer Phone
>>: (204)631-2298
>>Senior PWB Designer FAX :
>>(204)631-2475
>>NORTEL - Broadband Wireless Access email : [log in to unmask]
>>37 Stevenson Road
>>Winnipeg, MB R3H 0H9
>>
>>################################################################
>>DesignerCouncil E-Mail Forum provided as a free service by IPC using
>>LISTSERV
>>1.8c
>>################################################################
>>To subscribe/unsubscribe, send a message to [log in to unmask] with
>following
>>text in the body:
>>To subscribe: SUBSCRIBE DesignerCouncil <your full name>
>>To unsubscribe: SIGNOFF DesignerCouncil
>>################################################################
>>Please visit IPC's web site (http://www.ipc.org) "On-Line Services"
>section
>>for additional information.
>>For technical support contact Hugo Scaramuzza at [log in to unmask] or
>>847-509-9700 ext.312
>>################################################################
>>
>>################################################################
>>DesignerCouncil E-Mail Forum provided as a free service by IPC using
>LISTSERV 1.8c
>>################################################################
>>To subscribe/unsubscribe, send a message to [log in to unmask] with
>following text in the body:
>>To subscribe: SUBSCRIBE DesignerCouncil <your full name>
>>To unsubscribe: SIGNOFF DesignerCouncil
>>################################################################
>>Please visit IPC's web site (http://www.ipc.org) "On-Line Services"
>section for additional information.
>>For technical support contact Hugo Scaramuzza at [log in to unmask] or
>847-509-9700 ext.312
>>################################################################
>>
>>
>>
>____________________________________
>
> PCB Designs by Baldwin Tech.
>(602) 539-4400 FAX (602) 539-4500
> http://www.neta.com/~btech
>____________________________________
>
>################################################################
>DesignerCouncil E-Mail Forum provided as a free service by IPC using
>LISTSERV 1.8c
>################################################################
>To subscribe/unsubscribe, send a message to [log in to unmask] with
>following text in the body:
>To subscribe: SUBSCRIBE DesignerCouncil <your full name>
>To unsubscribe: SIGNOFF DesignerCouncil
>################################################################
>Please visit IPC's web site (http://www.ipc.org) "On-Line Services"
>section for additional information.
>For technical support contact Hugo Scaramuzza at [log in to unmask] or
>847-509-9700 ext.312
>################################################################
>
>################################################################
>DesignerCouncil E-Mail Forum provided as a free service by IPC using
LISTSERV 1.8c
>################################################################
>To subscribe/unsubscribe, send a message to [log in to unmask] with
following text in the body:
>To subscribe: SUBSCRIBE DesignerCouncil <your full name>
>To unsubscribe: SIGNOFF DesignerCouncil
>################################################################
>Please visit IPC's web site (http://www.ipc.org) "On-Line Services"
section for additional information.
>For technical support contact Hugo Scaramuzza at [log in to unmask] or
847-509-9700 ext.312
>################################################################
>
>
>
____________________________________
PCB Designs by Baldwin Tech.
(602) 539-4400 FAX (602) 539-4500
http://www.neta.com/~btech
____________________________________
################################################################
DesignerCouncil E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe: SUBSCRIBE DesignerCouncil <your full name>
To unsubscribe: SIGNOFF DesignerCouncil
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################
|
