RE: [SI-LIST] : AC Coupling vs DC Coupling

About this list Date view Thread view Subject view Author view

From: S. Weir (weirsp@a.crl.com)
Date: Mon Apr 17 2000 - 14:35:16 PDT


Farrokh,

What you are missing are the group delay effects of the transmission
line. Spreading the frequency content over a wide range results in
different relative arrival times for the transitions that define the bit
cell boundaries, ie jitter. Whether or not this is a problem is a matter
of coefficients. As you note, the 1394 standard uses 8b/10b coding to
limit the frequency content of the data to a fairly narrow range of less
than 3:1.

Regards,

Steve.
At 12:36 PM 4/17/2000 -0700, you wrote:
>I was under the impression that AC coupling is mainly used in cabled
>environments where we are trying to avoid common mode offsets and noise
>terms and there is concern for ground loops. Is this correct? If so, and
>if your signalling is contained to the backplane, DC coupling is a good
>choice. Also, assuming that you are using a start/stop protocol (and not
>using a continuous PHY like Fibre Channel), getting stable voltage levels
>and signals through the capacitor could take many microseconds. For these
>and other reasons (galvanic isolation), the 1394 bus standard defined AC
>coupling with 8b/10b for their S800 cabled topology while retaining the DC
>coupled backplane profile.
>
>Regards,
>
>Farrokh Mottahedin
>
>Quantum Corp.
>500 McCarthy Blvd.
>Milpitas, CA 95035
>(408)324-7934
>farrokh.mottahedin@quantum.com
>
>
>-----Original Message-----
>From: Bryan Robb [mailto:brobb@lsil.com]
>Sent: Monday, April 17, 2000 9:45 AM
>To: si-list@silab.eng.sun.com
>Subject: [SI-LIST] : AC Coupling vs DC Coupling
>
>
>
>Hi,
>
>I am trying to determine the pros and cons of using
>AC vs DC coupling in a backplane application. From looking
>at several different types of systems, it appears as if
>both are used in some form or another. I am thinking
>about DC coupling the system, but I don't want to overlook
>any compelling reason to AC couple.
>
>Furthermore, I am trying to maximize my data transfer and
>do not want to incur the penalty of an encoding scheme like
>8b/10b. So if I don't encode the data stream in some manner,
>the data will not be DC balanced which would automatically rule
>out AC coupling. Would non-DC balanced data cause a problem in a
>system that is DC Coupled? I am speaking about data transfer
>rates in the 622 to 800 MHz range.
>
>Thanks in advance,
>
>Bryan
>
>--
>****************************************************************************
>*
> Bryan Robb LSI Logic Corporation
> Engineering Manager 200 West Street
> Waltham, MA 02154
> Phone (781) 290-0238
> Fax (781) 890-6158
> Pager (888) 524-9468
>
> email: brobb@lsil.com
>****************************************************************************
>*
>
>**** To unsubscribe from si-list or si-list-digest: send e-mail to
>majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list
>or UNSUBSCRIBE si-list-digest, for more help, put HELP.
>si-list archives are accessible at http://www.qsl.net/wb6tpu
>****
>
>**** To unsubscribe from si-list or si-list-digest: send e-mail to
>majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
>si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
>si-list archives are accessible at http://www.qsl.net/wb6tpu
>****

**** To unsubscribe from si-list or si-list-digest: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:36:13 PDT