[SI-LIST] Re: AW: Re: Uncoupled differential signalling
- From: "Lee " <leeritchey@xxxxxxxxxxxxx>
- To: <Gert.Havermann@xxxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Mon, 1 Dec 2014 08:40:56 -0800
There is nothing intrinsically wrong with routing members of a diff pair on
different layers as long at the noise on the planes over which they route is
similar.
Differences in via barrel length are in the noise as far as length
mismatches are concerned. Differences in flight time due to weave
variations is much more important and that can happen even when they are
routed in the same layer and are tightly coupled. This issue should take up
more of our time than it does.
At very high data rates, whether the signal travels the length of a via or
just part of it can have a very big impact on signal quality, even when both
members are routed on the same layer.
-----Original Message-----
From: Havermann, Gert
Sent: Monday, December 01, 2014 3:52 AM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] AW: Re: Uncoupled differential signalling
At least single ended routing on different Layers is a bad idea. Broadside
coupled differential routing can be a good solution in some applications. It
also suffers from Layer thickness variations and via length differences, but
besides that it is much more stable and noise immune than single ended
routing.
But for most of the applications differential Stripline is the best way to
go.
BR
Gert
----------------------------------------
Absender ist HARTING Electronics GmbH, Marienwerderstraße 3, D-32339
Espelkamp; Registergericht: Amtsgericht Bad Oeynhausen; Register-Nr.: HRB
8808; Vertretungsberechtigte Geschäftsführer: Dipl.-Kfm. Edgar-Peter Düning,
Dipl.-Ing. Torsten Ratzmann, Dipl.-Wirtschaftsing. Ralf Martin Klein
-----Ursprüngliche Nachricht-----
Von: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] Im
Auftrag von Johan Lans
Gesendet: Montag, 1. Dezember 2014 11:03
An: si-list@xxxxxxxxxxxxx
Betreff: [SI-LIST] Re: Uncoupled differential signalling
Ok, diffrouting on separate layers seems to be a bad idea then. Thanks for
all the great answers!
2014-12-01 9:12 GMT+01:00 Havermann, Gert <Gert.Havermann@xxxxxxxxxxx>:
> Johan,
>
> routing differential pairs as single ended traces takes much more
> space than routing differential, because you need distance to isolate the
> traces.
> Routing on different Layers might look like a nice alternative as you
> can obviously copy and paste the routing, but that’s not true for high
> speed signaling. You will have different via barrel length, and via
> stub length along with the potential difference in Material
> properties, and that leads to timing problems if net well compensated.
> You also loose much of the noise immunity that differential signaling
> has been invented for because the noise will be very different on these
> different Layers.
>
> If the differential traces are too wide, just lower the substrate
> thickness until you reach a good routability. This is especially
> helpful when you are forced to route thru many dense via fields. The
> additional Losses are acceptable.
>
> BR
> Gert
>
>
> ----------------------------------------
> Absender ist HARTING Electronics GmbH, Marienwerderstraße 3, D-32339
> Espelkamp; Registergericht: Amtsgericht Bad Oeynhausen; Register-Nr.:
> HRB 8808; Vertretungsberechtigte Geschäftsführer: Dipl.-Kfm.
> Edgar-Peter Düning, Dipl.-Ing. Torsten Ratzmann, Dipl.-Wirtschaftsing.
> Ralf Martin Klein
>
> -----Ursprüngliche Nachricht-----
> Von: si-list-bounce@xxxxxxxxxxxxx
> [mailto:si-list-bounce@xxxxxxxxxxxxx]
> Im Auftrag von Scott McMorrow
> Gesendet: Samstag, 29. November 2014 19:00
> An: johan.lans@xxxxxxxxxxxxxxx
> Cc: si-list@xxxxxxxxxxxxx
> Betreff: [SI-LIST] Re: Uncoupled differential signalling
>
> Johan,
> There are 2 issues. One that you have brought up yourself, which is
> that the noise environment in different planar cavities will be
> different, causing a noise imbalance on the pair. I imagine that this
> could be managed with Gnd-Gnd cavities and appropriate stitching. The
> second issue is a difference in the dielectric environment, due to the
> anisotropic properties of the laminate, which will cause unpredictable
> skew. This is probably the more difficult problem to resolve.
>
> Scott
>
>
>
> Scott McMorrow
> Teraspeed® Consulting - A Division of Samtec
> 16 Stormy Brook Rd
> Falmouth, ME 04105
> (401) 284-1827 Business
> http://www.teraspeed.com
>
> On Sat, Nov 29, 2014 at 2:42 AM, Johan Lans
> <johan.lans@xxxxxxxxxxxxxxx>
> wrote:
>
> > On a recent design I tried out routing loosely coupled differential
> pairs.
> > Previous tests I've have done showed that loosely coupled pairs are
> > more consistent in differential impedance and easier to simulate
> > accurately, so I think that there are good reasons to try this out.
> > However, routing proved kind of tricky since the total width of the
> > pair was large. A colleague suggested routing the traces as
> > uncoupled 50ohm traces. If this is done on layers that are separated
> > by one or more planes, one could route the traces almost exactly
> > identical, over each other (perhaps compensating for the two signals
> > traversing different lengths in vias).
> > Would there be any negative effects to doing this? My thought is
> > that between different planes in the stackup, there could be
> > different noise affecting the transmission lines so that the noise
> > cancelling effect of the differential signalling is without effect.
> > /Johan
> >
> >
> > ------------------------------------------------------------------
> > 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 forum is accessible at:
> > http://tech.groups.yahoo.com/group/si-list
> >
> > List archives are viewable at:
> > //www.freelists.org/archives/si-list
> >
> > 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:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>
> List forum is accessible at:
> http://tech.groups.yahoo.com/group/si-list
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
>
> 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:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
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:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2015.0.5577 / Virus Database: 4223/8660 - Release Date: 11/30/14
------------------------------------------------------------------
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 forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
