[SI-LIST] Re: Routing 10G differential lines over Standard FR-4
- From: Bill Wurst <billw@xxxxxxxxxxx>
- To: si-list@xxxxxxxxxxxxx
- Date: Mon, 16 May 2005 09:02:58 -0400
Sunil,
Unfortunately, there are no simple answers at 10Gbps. At that speed,
there are three main concerns and you have already mentioned two:
dielectric losses and discontinuities. The third is skin-effect losses,
which can be mitigated by widening your traces and spaces and increasing
the dielectric thickness to maintain a 100-Ohm differential impedance.
To be certain of the design, one must develop a link budget and
determine what margin remains after all losses (including reflections
from discontinuities) have been accounted for. If your link length is
short, then it may be possible to ignore dielectric and skin-effect
losses and use standard FR-4. What defines "short" will depend on your
design; the dividing line may be on the order of 10cm at this speed.
As you point out, the advantage of microstrip is that no vias are
required. The disadvantages are that the plating process on the outer
layers makes it more difficult to control the impedance, and soldermask
tends to be fairly lossy. Fortunately, at 10Gbps non-TEM modes that
exist in non-homogeneous dielectrics should not be a problem yet.
Stripline can be made to work, but attention must be paid to both the
signal vias and the vias in the reference planes for the return current,
including the signal via stubs (blind vias or back-drilling can
eliminate these). A 3D field solver can be most helpful here.
Some useful references:
High-Speed Signal Propagation by Johnson/Graham (particularly
Chapters 2, 3 and 6, as well as section 5.5 on vias),
A SI Introduction to RocketIO Transceivers, and
Loss Budgeting for RocketIO Transceivers.
The last two are DVDs available from Xilinx at:
http://www.xilinx.com/products/design_resources/signal_integrity/index.htm
and contain much useful information no matter which vendor's parts you
are using. Scroll to the bottom of the page for more information.
Again, no simple answers but I hope I've been helpful.
-Bill
/************************************
/ billw@xxxxxxxxxxx /
/ /
/ Advanced Electronic Concepts, LLC /
/ www.aec-lab.com /
************************************
=================================================
sunil.mekad@xxxxxxxxx wrote:
> Hi Everyone,
>
> I was planning to route high speed differential traces 10Gbps over
> standard FR-4 material. Can anyone suggest which FR-4 material can be
> used (Nelco N4000-6 or Rogers RO4350)? Is it possible to do use standard
> FR-4 without compromising signal quality.
>
> Also is it recommended to route these traces over microstrip rather than
> striplines traces? If I route it over stripline then it would require
> vias to be put on every pair of these differential traces .. I am not
> too comfortable doing that!!
>
> Can anyone tell what can be done? Thanks in anticipation,
>
> Regards,
> Sunil
>
>
>
> Confidentiality Notice=0D
>
> The information contained in this electronic message and any attachments to=
> this message are intended
> for the exclusive use of the addressee(s) and may contain confidential or=
> privileged information. If
> you are not the intended recipient, please notify the sender at Wipro or=
> Mailadmin@xxxxxxxxx immediately
> and destroy all copies of this message and any attachments.
> ------------------------------------------------------------------
------------------------------------------------------------------
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 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:
//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:
