FreeLists / si-list / [SI-LIST] Re: Differential microstrip with coplanar ground traces... unexpected results

["Scott McMorrow" <scott@xxxxxxxxxxxxx>]

Bob,

You are absolutely correct.  Coplanar waveguide can be extremely 
twitchy, especially with narrow ground traces.  You generally need to 
have much wider grounds in order to better contain the fields. 
 Otherwise, crosstalk from each side will cause some nasty mode 
conversions and crosstalk to occur. In general, CPW or grounded CPW, as 
your case is, is useful for better impedance control and lower loss only 
when you can use a wider ground line.  In addition, if your edge rates a 
fairly fast, you may begin to excited a half-wave resonance between your 
vias.  For your configuration with vias spaced at 500mils, you can 
expect a resonance point at around 6 to 7 GHz on Fr-4, depending on how 
far away the ground plane is from the microstrip layer.  At the 
resonance point, the crosstalk will become extremely high. This would be 
right around the 5th harmonic of your 2.5 Gbps signal.

I use Ansoft 2D to evaluate configurations like this, and treat the 
ground traces as signal traces in the modeling.  Then with 3D, you can 
extract the via model and perform a complete simulation of the signal 
and ground traces.  I;ve had success with 10 Gbps signals in a package, 
but in that case, we had ample room to form a good ground plane between 
the pairs.

best regards,

scott

-- 
Scott McMorrow
Teraspeed Consulting Group LLC
2926 SE Yamhill St.
Portland, OR 97214
(503) 239-5536
http://www.teraspeed.com



Bob Welte wrote:

>Hello to All
>
>Would anybody like to share their knowlege or experience with using
>coplanar ground traces in a microstrip configuration?
>Are there any "rules" for the width or impedance, and number/spacing of
>vias for the ground traces of such a configuration?
>
>A short explaination of our design:
>We have 100 ohm diff microstrip; signals are .013", with .015" spacing,
>ground traces are spaced .010" from signals and are .010" wide, with vias
>to the ground plane (.012"
>below) at about every .5".  Our digital signals are 2.5Gb/s with the
>fastest rise time at 100ps.
>
>We expected the coplanar grounds to better contain the fields, and allow
>closer pair to pair spacing without the risk of large crosstalk.  We
>believed that we had taken the coplanar grounds into account when designing
>the configuration, but now thnk that the software we used assumed that the
>coplanar grounds were perfect.  With the narrow ground traces, and the vias
>to the ground plane, we now believe that the coplanar grounds are actually
>quite inductive at our frequencies.  We have both microstrip and stripline
>coupons, and expected the microstrips to be slightly better since they have
>no vias in the signal path.  But in fact the striplines show less
>discontinuity (TDR), and have less jitter when transporting a pseudorandom
>bitstream.
>Thanks
>Bob Welte
>IBM Microelectronics
>
>
>
>
>------------------------------------------------------------------
>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 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 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