[SI-LIST] Re: Diff.Pairs

  • From: Michael Chin <mchin@xxxxxxxxx>
  • To: MikonCons@xxxxxxx
  • Date: Mon, 13 Oct 2003 18:29:18 -0700

Mike,

I fully agree with your comments on the tightly coupled traces and
your interpretation of Lee's comments.  When Lee mentioned about
pin escaping for differential signals in a 1mm pitch BGA, my mind set
has already switched to the broadside coupling.  This is the only way to
take the diff. signals out of a 1mm BGA, and continue outside with
a coplanar differential signals (at the cost of 1 extra VIA, and another
layer).

I always has a question that may be you can shed some light:

"Given that we have to use broadside coupling routing to exit the pin array,
  is it better to continue this broadside coupling, or is it better to switch 
to the
  coplanar route when have a chance at a cost of an VIA ?  To put some limits
  on the application, lets assume that this is for a 622MHz LVDS, running 5" to
  the destination ASIC."

Thanks,
Michael

MikonCons@xxxxxxx wrote:

> Hi, Chris:
> I admired your comments and the associated question you posed. If you caught
> my comment on this subject sent a few minutes ago, you will see that there is
> no way I would split pairs to different layers (unless forced into using
> broadside coupled pairs). My interpretation of Lee's comments was that he
> occasionally uses the traces of a differential pair on totally separate 
> layers (i.e.,
> as separate 50-Ohm traces). There was no mention (that I caught) of
> broadside-coupled pairs, which is a workable solution, but has other design 
> tradeoffs.
> Lee Richey and I do not appear to be on the same page.
>
> Regarding "tightly coupled" traces:  The degree of coupling tends to fall in
> the 10-20% range; therefore, there is still considerable current implied in
> the nearest image plane (but of course, with opposite polarity so the net
> current is very small). I use tightly coupled traces with reasonable 
> separation from
> adjacent pairs to a) minimize the net differential signal coupled into any
> given pair, and b) minimize impedance variations caused by manufacturing
> tolerances in the dielectric materials.
>
> Mike
>
> Michael L. Conn
> Owner/Principal Consultant
> Mikon Consulting
>

------------------------------------------------------------------
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 archives are viewable at:     
                //www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu
  

Other related posts: