[SI-LIST] Re: 2 questions about TRL cal
- From: Jim Nadolny <jim.nadolny@xxxxxxxxxx>
- To: Mick zhou <mick.zhou@xxxxxxxxx>, "si-list@xxxxxxxxxxxxx" <si-list@xxxxxxxxxxxxx>
- Date: Mon, 8 Aug 2011 14:45:09 +0000
Hey Mick -
My 2 cents...
On the delays - I always measure the delay differences in the line standards
first, then specify the delays in the cal kit definition. The result has been
pretty good success with the behavior of the extracted models. 4 or 5 years
ago I was starting to dig into all this and was told (by an industry colleague)
that the Agilent code does not use the delay numbers...so just use 0 ps. When
I asked Agilent about that practice they sort of looked at me like I had 3
heads - yes you need to specify the delay so that is what I do. To be honest -
I can't really answer your question though, just my experience.
On the overlap - I use the overlap as a safety measure to cover variations in
substrate performance. Gives me margin when dealing with batch to batch
variations in FR-406. Having the overlap also allows you a range of
frequencies that you can use to intelligently select as a transition frequency.
The selection of the transition frequency can impact the "steps" in the
measured RL that is characteristic of TRL/M calibration.
jn
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
Behalf Of Mick zhou
Sent: Monday, August 08, 2011 10:19 AM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] 2 questions about TRL cal
All,
There are two simple questions about TRL calibration that papers/app notes are
not very clear about.
1. Is it necessary to enter delays of lines?
The entries are available in the GUI. I have seen people think the delays
are not necessary and use 0ps for all. The argument is: the delay is determined
as part of the process. According to the theory, S12 and
exp(-gamma*l) are determined to a sign ambiguity. Extra infomation such as
delay is needed to uniquely decide the sign (therefore the phase). However,
delay may not be a problem for engineers who don't care phases. because they
don't "see" the difference by looking at magnitudes only. But this could be
dangerous if the models without correct phases are used in time domain
simulations, the delays could be wrong unless you are lucky the phases are
always within 20-160 degrees (8:1 case).
2. Is overlap of frequencies necessary in case of multi-lines? Again it is
related to accuracy of phases. Some believe overlaps can improve phase
accuracy. Some don't believe so. The instrument picks up either the last
(unguided cal) or the prority set by Smartcal (guided cal). So how the phase
accuracy is improved?
Thanks
Mick
------------------------------------------------------------------
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
------------------------------------------------------------------
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:
