[SI-LIST] Re: single RLC parameters at a certain frequency
- From: "Eric Bogatin" <eric@xxxxxxxxxxxxxxx>
- To: "'xuzhengrong'" <Zhengrong.Xu@xxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Tue, 20 May 2008 10:01:25 -0500
Hi folks-
There is a difference between the bandwidth of a model, the bandwidth of a
measurement and the bandwidth of an interconnect.
If you use a single section RLC model to model an interconnect, the model
will have a limited bandwidth, based on the length of the interconnect. The
bandwidth of the model is roughly when the interconnect length is 1/10th a
wavelength. You don¡¯t have to do a measurement up to this frequency to know
the single section RLC model will not work beyond this 1/10th wavelength
frequency.
As a rough rule of thumb, the bandwidth of an RLC model is about BW = 0.6
GHz/Len(inches), where Len is the length of the interconnect, in inches,
assuming an FR4 like interconnect.
For example, if the package lead length is 0.25 inches, the bandwidth of the
single section RLC model would be about 2.5 GHz. It can be extracted from
data measured at low frequency, and still used at much higher bandwidth. You
can fit the RLC model to S-parameter data at any frequencies before this
bandwidth.
If you try to fit an RLC model to S-parameter data above the bandwidth of
the RLC model, you can probably fit one frequency, but not down to DC. You
need a more complex model.
If you use a T element model, rather than an RLC model, it can often have a
bandwidth higher than the 1/10th wavelength limit, if the interconnect is
uniform. In this case, the bandwidth of the model is often limited to when
the losses begin to dominate, which can be in the GHz regime. There is no
way to tell the bandwidth of the model, looking at low frequency data.
The only way of knowing the bandwidth of a model is performing a measurement
of the real structure with a measurement bandwidth beyond the model's
bandwidth. If you fit a T element model or a lossy T element model to 1 GHz,
for example, the model may work great to 5 GHz, you just can¡¯t tell unless
you look at measured results to 5 GHz.
--eric
*******************************************************
Eric Bogatin
Signal Integrity Evangelist
Bogatin Enterprises
Setting the Standard for Signal Integrity Training
26235 W 110th Terr
Olathe, KS 66061
e: eric@xxxxxxxxxxxxxxx
v: 913-393-1305
cell: 913-424-4333
f: 913-393-0929
www.BeTheSignal.com
San Diego: EPSI, BBDP, July 28-31, 2008
San Jose, SICT, Aug 12-13
San Jose, EPSI, BBDP, Sept 29-Oct 2
***********************************************
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
Behalf Of xuzhengrong
Sent: Monday, May 19, 2008 10:43 PM
To: si-list@xxxxxxxxxxxxx
Cc: 'ÑϺ½'
Subject: [SI-LIST] single RLC parameters at a certain frequency
Hi,
To simplify the model, we often extract single RLGC parameters from the
s-parameter model at a certain frequency.
I wonder what on earth the meaning of the single frequency parameters is.
First, if the parameters only represent the magnitude and phase of the
S-parameter model at the frequency point, it may have different combinations
of RLGC to reach the same goal, what's the method or principle?
Second, if the parameters are meaningful at a certain frequency, why can
single RLGC parameters often be used to estimate characteristic of package
model such as IBIS?
In my opinion, for example, the single RLGC model should be the first-order
approximation to the real model for transmission line, as Eric's book
'Signal Integrity: Simplified' said.
It is equivalent to the real model within low frequency range and we can use
it if the signal frequency component is within the range.
However, if the RLGC model extracted at a certain frequency is only
meaningful for the extracting frequency, it isn't the first-order
approximation for the transmission line.
So we can't use it as the first-order approximation and estimate its low
frequency characteristic.
Appreciate any idea and discussion for the problem.
Best Regards,
Zhengrong
------------------------------------------------------------------
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
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
------------------------------------------------------------------
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
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:
