[SI-LIST] Re: frquency limit of a channel
- From: Istvan Novak <istvan.novak@xxxxxxx>
- To: "Lakshmi N. Sundararajan - PTU" <lakshmi.s@xxxxxxxxxxx>
- Date: Sat, 06 Mar 2010 23:30:05 -0500
Hi Lakshmi,
As always, it depends. If you have a well-behaved channel, with very
little crosstalk and SSN,
smooth S21 curve and low reflections (also well over 3GHz), in that
case the frequency response
beyond 3GHz will have little impact on the overall behavior. The
flipside is: how do we know
that these parameters are well behaved beyond 3GHz unless we model them
and look at them
in the first place?
At high frequencies we may have many structural resonances from various
sources: via stubs,
trace/package-pin stubs, arrays of features like vias and their
pads/antipads. Resonance-like
phenomena tend to be narrow band, but they can have a significant impact
within their
characteristic band. Imagine a resonating structure (could be a
trace-stub or power-puddle
resonance), creating a suckout above 3GHz (via stubs in typical boards
will have much higher
frequencies). The suckout will have an impact on the worst-case eye
opening, either horizontally,
or vertically or both.
There is an even more important argument for extending the channel
modeling way beyond
the bit-rate frequency, at least in high-performance systems, which
always push the limit
in density and performance: crosstalk, especially when coupled with
structural resonances, can
BOOST signal, not only attenuate. And if this happens close to the
receiver, the only thing
protecting us against this noise is the sensitivity roll-off of the
receiver; the smooth channel
transfer function wont help.
Regards,
Istvan Novak
Oracle
Lakshmi N. Sundararajan - PTU wrote:
> Hi Gurus,
> Suppose assume I have a high speed serial link at 6Gbps. The nominal
> rise time of the signals on this channel is 150ps.
>
> Given this rise time, the bandwidth required to transmit this signal is
> 0.35/tr = 2.33Ghz.
>
>
>
> So, to study this channel behavior, is it correct to only look at
> s-param frequency output till say 3Ghz.
> Can any higher frequency data points on this s-param be ignored and
> still correctly model the channel behavior?
>
>
>
> I also looked up BW * tr = 0.35. This equation is derived from a simple
> RC integrator circuit.
>
> How true can this model any channel, since we seem to be using this
> equation for all our studies.
>
> Please clarify.
>
>
> Thanks,
>
> -LN
>
>
>
>
>
>
>
>
>
>
------------------------------------------------------------------
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:
