[SI-LIST] Re: differential pair routing topology

Hi Andy,
Thanks for your reply! At 1.65GHz max (DVI video) the wavelength is only 
a few inches. Most of the trace runs on this board are not longer that 
that. Requirements for precisely maintained pair gaps and 45 degree 
corners only would be overkill. I think the more important things are:
1) The pair members are matched in length, and to a lesser extent the 
number of corners.
2) Other signals are kept far enough away to avoid crosstalk problems.
3) The stripline return paths (power & ground) are clean (no splits, 
good bypassing, etc).
Anyone disagreeing, or with more comments, please let me know.
Sincerely,
Ivor.

on 4/24/2006 6:41 AM PST Andrew Ingraham wrote:
>> 1) I'm considering the routing topology. Density is a factor, space for
>> routing channels is limited. For width / clearance rules I could use,
>> say, 4/8/4 (mils). I think this could be considered "loosely coupled".
>> This could be routed using a 12 (or divisor of) mil grid. But this
>> spacing can not be maintained the entire length, it must vary slightly
>> at the IC and connector fanouts / vias. So the question is, would it be
>> better to increase the spacing to maintain a more consistent impedance
>> profile, or does it not make a significant difference?
> 
> Consider the length of the impedance discontinuity with respect to the
> shortest wavelength.  Maybe it doesn't make enough difference.
> 
> Decreasing the spacing is similar to allowing traces of other signals nearby
> in that region.  Some people get anal trying to maintain precisely some
> particular diff. pair spacing rule, while allowing other traces to impinge
> on the area.
> 
>> 2) I understand the primary reason for not using 90 degree corners is
>> the impedance discontinuity due to the capacitance of the extra corner
>> area, so it is better to use 45 degree corners.
> 
> Actually, that's not it.  For most of us who work with frequencies below
> 10 GHz or more, the reason for avoiding 90 degree corners is not an
> electrical one.  It is to improve yield, since sharp corners tend to be
> somewhat problematic for good etching.
> 
> The electrical discontinuity is there but it's very slight and hardly
> matters unless you're way up into the microwaves.  But there's a lot of
> folklore that ignores this fact.  Some also claim corners become antennas,
> which may be true but only if you go really high in frequency.
> 
> Check with your board vendor about using non-chamfered (90 degree) corners.
> 45 degree bends are generally better but not always necessary.  Consider
> also whether you're running significant lengths of traces at 45 degrees, or
> only tiny segments.
> 
> Regards,
> Andy
> 
> 
> ------------------------------------------------------------------
> 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:
> http://www.freelists.org/webpage/si-list
> 
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
> 
> List FAQ wiki page is located at:
>                 http://si-list.org/wiki/wiki.pl?Si-List_FAQ
> 
> List technical documents are available at:
>                 http://www.si-list.org
> 
> List archives are viewable at:     
>               http://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
>   
> 
> 
> 

------------------------------------------------------------------
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:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List FAQ wiki page is located at:
                http://si-list.org/wiki/wiki.pl?Si-List_FAQ

List technical documents are available at:
                http://www.si-list.org

List archives are viewable at:     
                http://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: