If this is not the appropiate place, forgive me all.
We are looking for a person who would come and work designing high speed
boards and interface with various groups, sales, engineering,
If you want to grow in a small atmosphere, please send me your resume or
3dfx interactive Inc
4435 Fortran Dr.
San Jose, CA 95134
>From: Ray Anderson[SMTP:raymonda@radium.Eng.Sun.COM]
>Reply To: Ray Anderson
>Sent: Monday, November 10, 1997 4:11 PM
>Subject: re: [SI-LIST] : Guard banding
> The context in which you usually hear about
> "guard traces" is in very high impedance
> analog circuitry where the designer is
> trying to guard a very hi-Z node from
> very minute leakage currents flowing across
> a board or substrate. In these cases, a grounded
> guard trace can be effective.
> I tend to agree with Fabrizio, that in
> high speed circuitry where one is concerned
> with AC crosstalk, that the use of grounded
> guard traces may be of dubious value.
> So depending on what Ronda was referring
> to when she asked about grounded guard traces,
> the answer could go either way.
>> Ronda, I don't have any papers to supply to you, but I
>will mention one
>> thing about the subject. In my experience, I have never
>found guard bands
>> to be a worthwhile thing to add to a design. It adds
>traces, makes the
>> board more complex, and the distance you need to space
>critical traces from
>> each other to fit the guard band is enough to eliminate
>> the traces. Guard bands also lower the impedance of
>these critical nets.
>> So unless one is running 2GHz+ signals, single ended on a
>PC board, I would
>> recommend not using guard bands.
>> Fabrizio Zanella
> Sr. Design Engineer
> Hardware Engineering
>> _/ _/
>> _/_/_/_/ _/ _/ _/_/_/ _/ voice:
>(508) 435-1000 Ext.
>> _/ _/_/ _/_/ _/ _/ Fax:
>> _/_/_/_/ _/ _/ _/ _/ _/_/_/
>> _/ _/ _/ _/ EMC
>> _/_/_/_/ _/ _/ _/_/_/ 171
>Hopkinton, MA 01748-9103