TECHNET Archives

1996

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Condense Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Received:
by ipc.org (Smail3.1.28.1 #2) id m0ueKwu-0000EjC; Thu, 11 Jul 96 07:34 CDT
Content-Type:
text/plain; charset=us-ascii
CC:
References:
Date:
Thu, 11 Jul 1996 08:42:43 -0400
Precedence:
list
Resent-From:
Resent-Sender:
TechNet-request [log in to unmask]
MIME-Version:
1.0
Status:
O
X-Mailing-List:
<[log in to unmask]> archive/latest/5081
TO:
"Linahan, Andrew" <[log in to unmask]>
Return-Path:
<TechNet-request>
X-Status:
X-Loop:
Old-Return-Path:
Content-Transfer-Encoding:
7bit
Resent-Message-ID:
<"Q_mJy3.0.gxD.JJFvn"@ipc>
Subject:
From:
[log in to unmask] (Doug McKean)
From [log in to unmask] Thu Jul 11 13:
33:21 1996
X-Mailer:
Mozilla 2.0 (Win95; I)
Organization:
Paragon Networks Inc.
Message-ID:
Parts/Attachments:
text/plain (82 lines)
Linahan, Andrew wrote:
> 
> Dear Technet,
> 
> Can anyone direct me to the results of any studies into the relative
> effectiveness of "encapsulating" all PCB traces between outer layer
> ground planes (of multilayer PCBs) to minimise electromagnetic emissions?
> We are contemplating adding this feature to designs which are about to
> enter CAD, but have yet to be convinced that the (routing) "pain" is
> worth the "gain". Failing that, a summary of anyone's practical
> experience with this or any other (EMC-enhancing) technique would be
> greatly appreciated.
> 
> Andrew Linahan
> 
> (Manufacturing Test Manager)
> QPSX Communications Pty Ltd
> Private bag 24
> West Perth 6005
> Western Australia
> 
> tel. +61 (09) 262 2000
> fax. +61 (09) 324 1642
> email [log in to unmask]
> 
> ***************************************************************************
> * TechNet mail list is provided as a service by IPC using SmartList v3.05 *
> ***************************************************************************
> * To unsubscribe from this list at any time, send a message to:           *
> * [log in to unmask] with <subject: unsubscribe> and no text.        *
> ***************************************************************************

Andrew,

Actual measurements from one design to another I don't have.
If you want to start a discussion about what you are doing, 
I'd be glad to that ON-LINE or OFF-LINE.  

My experience covers the KHz to GHz, motor servo control to
communications circuits, TTL to ECL, impedence control to 
no impedence control.  Although it may sound like alot,
it's really not compared to some other people I know.

The "encapsulation" method may or may not work depending 
upon alot of factors.  There are some questions we'd have to 
work through so that I can get and idea of what you're up against:

1.  What are some of the oscillators on board?
2.  What are your edge rates?
3.  Are we talking an impedence controlled board?
4.  Does the board have very little interface with other
    boards or outside world, or are there alot of interfaces to
    other boards and the outside world?
5.  What family or families of devices are on the board?
6.  What is the purpose of the board?
    Is it controlling a little motor in a machine shop?
    Is it controlling 1200 emergency phone calls a day?
    Is it controlling patient connected devices in a hospital?

Although emi control at the pcb level is where any design should start,
it happens to be the most obscure area of emi control with designers.

There are a couple of books published on the subject.
I can supply a list of them for you.

I can also provide some rough "rules-of-thumb" to use for consideration.

I could go on but I'd be taking up time.  Give me a shout!

Doug

[log in to unmask]

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************



ATOM RSS1 RSS2