[SI-LIST] Re: Routing differential lines as single ended traces ?
- From: "Ken Cantrell" <Ken.Cantrell@xxxxxxxxxxxxxxxx>
- To: "Danny Damhave" <dd@xxxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Tue, 27 Sep 2011 09:22:26 -0600
Danny,
Agreed that with your material choice you would lose routing density.
However there are other material options available.
Material Thickness
1080 - 2.5 mils
2116 - 3.8 mils
1086 - 2.0 mils
1067 - 1.3 mils
1080 is 60x47
2116 is 60x58
1086 is 60X60
1067 is 70x70
1086 and 1067 are square symmetric where the others are not, and would be
better choices than 2116. They also offer the advantage of being thinner,
so routing density would be improved over 1080 or 2116. With the correct
laminate choice for your particular application, it is possible to achieve
low/no laminate induced skew, and maintain low crosstalk and high routing
density.
Ken
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Danny Damhave
Sent: Monday, September 26, 2011 11:42 PM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Routing differential lines as single ended traces
?
Using eg 2116 instead of eg 1080 will increase thickness of the laminate,
thereby the hight of the traces above ref planes and thereby trace width
because Er for the structures width a different laminate is not increased so
much. The overall crosstralk on the PCB has as strong dep of the trace to
ref distance and strong dep of he trace to trace space. Therefore the cross
talk will increase or you will have to increase the trace to trace spacing
and have fewer routing channels.
BR
Danny
On 26/09/2011, at 23.47, asparky@xxxxxxxxxxxxxxxxxxxxxxxx wrote:
> Why do you think changing glass weaves increases crosstalk?
> Thanks,
> Aubrey
> Sent from my Verizon Wireless BlackBerry
>
> -----Original Message-----
> From: Danny Damhave <dd@xxxxxxxxxxx>
> Sender: si-list-bounce@xxxxxxxxxxxxx
> Date: Mon, 26 Sep 2011 21:59:31
> To: <si-list@xxxxxxxxxxxxx>
> Subject: [SI-LIST] Re: Routing differential lines as single ended traces ?
>
> Right, but then we get more crosstalk instead ;-)
> BR Danny Damhave
>
> On 26/09/2011, at 21.45, Lee Ritchey wrote:
>
>> If you use the right glass weave it is not necessary to be concerned
about skew. No need for tricky routing.
>>
>>
>> --------------------------------------------------
>> From: "Danny Damhave" <dd@xxxxxxxxxxx>
>> Sent: Monday, September 26, 2011 12:24 PM
>> To: <si-list@xxxxxxxxxxxxx>
>> Subject: [SI-LIST] Re: Routing differential lines as single ended traces
?
>>
>>> Hi,
>>> Routing a LVDS differential pair with a tightly coupling, is sensitive
to production tolerances/variation eg. etching factor/coating.
>>> Having a weak coupling or using the not closer than rule (single ended
traces) is less sensitive.
>>> If your traces are wide and your timing is very critical is probably
best to route with a very weak or weak coupling and match the trace to trace
spacing with the pitch of the glass fibres in your laminate to reduce the
skew between the signals, but of course this can also be achieved by staying
away from 90deg routing and using the not closer rule.
>>> I assume that your signals always have a reference plane, are not close
to any edge of the PCB, do not leave the board , and are not close to noise
sources like DC DC switchers and your signals are not influenced by PCB
resonances or the like.
>>> BR
>>> Danny Damhave
>>>
>>>
>>>
>>> On 26/09/2011, at 03.57, Low Jerry wrote:
>>>
>>>> Hi,
>>>> I recently came across a validation platform where the LVDS
differential
>>>> lines are routed as single ended traces traces instead of tightly
coupled
>>>> differential pairs. The motivation I found was so that they could use
each
>>>> of the pairs as single ended traces when needed as well. When I probed
>>>> deeper it seems like the person who proposed this scheme has left. So I
>>>> would like to seek help here on some clarification
>>>>
>>>> - Will a scheme like this impact the performance of the differential
pair
>>>> ? Since I understand that differential routing is more immune to noise.
>>>> - What are the considerations/feasiblity studies that need to be done
>>>> before implementing a scheme like this ?
>>>> - Since this is a validation platform what measures can be taken ensure
>>>> that the performance seen are similar to a production platform if the
>>>> production platform is routed in differential.
>>>>
>>>> Thanks in advance for the feedbacks. Have a great day.
>>>>
>>>>
>>>> ------------------------------------------------------------------
>>>> 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 technical documents are available at:
>>>> http://www.si-list.net
>>>>
>>>> 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 technical documents are available at:
>>> http://www.si-list.net
>>>
>>> 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 technical documents are available at:
> http://www.si-list.net
>
> 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 technical documents are available at:
> http://www.si-list.net
>
> 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 technical documents are available at:
http://www.si-list.net
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 technical documents are available at:
http://www.si-list.net
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:
