[SI-LIST] Re: More discussion on AC coupling capacitor
- From: Scott McMorrow <scott@xxxxxxxxxxxxx>
- To: emcesd@xxxxxxx
- Date: Mon, 24 Sep 2012 11:03:36 -0400
"Does placing the AC capacitor near the receiver really help SI?"
Let's dispel this "fact" once again. In a linear interconnect that is
well-matched, it does not matter where a DC blocking capacitor is placed.
What matters are potential reflective resonances that occur between
mismatched structures on the interconnect. Dies, packages, connectors,
vias, and capacitors tend to be the major impedance discontinuities in an
interconnect. It does not matter if dispersion happens before or after a
discontinuity, or on both sides of it. The only thing that matters is
keeping the discontinuities as small as possible. This can be done with DC
blocking capacitors by performing optimization of the transition in and out
of the capacitor with electromagnetic modeling tools. It is quite possible
to keep 0402 capacitors "flat" out to 10 GHz and 0201 capacitors "flat" out
to 20 GHz with good designs.
"Which is the better place for placing the decoupling capacitor, near the
transmitter or near the receiver?"
If the capacitor transition design is not well-matched, then "best"
placement depends upon how it interacts with other structures with
impedance discontinuities. This might be packages, drivers, receivers, or
connectors. It just depends which is the worst, how close the capacitor is
to the structure, and the amount of interconnect loss in between them.
Loss is good, since it will de-Q resonances and improve return loss in the
region of the discontinuities by 2 times the interconnect loss between
them. For this reason, I like loss line and switch cards, and prefer to
use low loss practices in backplanes. Thus the best place to place an DC
blocking capacitor is as far away as possible from other discontinuities.
Oftentimes the worst discontinuity is the receiver, due to excessive
capacitance. Placing a capacitor near the receiver can sometimes be
counterproductive.
Finally, what is the correct value ... it depends on the low frequency
limit of your serial encoding pattern. The capacitor must be large enough
to reduce low frequency wander and jitter problems.
Regards,
Scott
--
Scott McMorrow
Teraspeed Consulting Group LLC
121 North River Drive
Narragansett, RI 02882
(401) 284-1827 Business
(401) 284-1840 Fax
http://www.teraspeed.com
Teraspeed® is the registered service mark of
Teraspeed Consulting Group LLC
On Mon, Sep 24, 2012 at 3:05 AM, Tesla <emcesd@xxxxxxx> wrote:
> Hi, Experts
>
> In the last thread,Michael said "Placing them at the RX side generally
> helps to minimize impact of the impedance discontinuity given the
> degradation of edge rate from dispersion"
> Does place the AC capacitor near the receiver really help SI?
> Which is the better place for placing the decoupling capacitor, near the
> transmitter or near the receiver?
> How to choose the value of the decoupling capacitor? it is based on the
> data rate, code style or other things? can we calculate it manually?
>
> Thanks.
>
> Tesla.
> ------------------------------------------------------------------
> 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
Other related posts:
