[SI-LIST] Re: spatial resolution and effective rise time of VNA with TDR time-domain option
- From: "Dima Smolyansky" <Dima.Smolyansky@xxxxxxxxxxxxxx>
- To: "Istvan Novak" <istvan.novak@xxxxxxxxxxxxxxxx>,<si-list@xxxxxxxxxxxxx>
- Date: Wed, 30 Oct 2002 08:44:18 -0800
Hello, Istvan:
I respectfully disagree with your application of 0.35/BW rule to estimate
the rise time of a time domain option in a VNA. As you yourself pointed out,
the windowing and other processing affects the achievable rise time. As a
result, a 20 Ghz network analyzer does not even come close to providing 17ps
rise time as the 0.35/BW equation would implicate. As a matter of fact, as
far
as I know you will need to use a 40 Ghz VNA to get approximately 30 ps rise
time.
Now, if we try to apply the rule that Jan mentioned to the example above:
15cm/bandwidth. Let's use 40 Ghz as an example; the equation seem to suggest
that we can resolve 0.375cm. In vacuum (Er=1) this would equal approximately
12.5ps, if my math does not fail me. That is just under 1/2 the rise time;
seems like the equation works in this case at least.
Having referred to the 1/2 the rise time being the figure of merit for
resolution, I should mention that it applies to the effective rise time;
that is, the actual rise time that can be slowed down by the losses in the
DUT by the time it reaches the point of interest. In addition, there are
well developed comparative techniques, commonly used with TDR oscilloscopes,
which allow one to resolve things which are much smaller than half the rise
time of TDR.
Finally, not to reject the rule of 0.35/BW; it is, of course, known to be a
good estimate which allows one to convert the rise time of a Gaussian or
almost Gaussian digital signal into its 3dB bandwidth and vice versa.
Thanks,
===================
Dima Smolyansky
TDA Systems, Inc.
11140 SW Barbur Blvd., Suite 100
Portland, OR 97219
(503) 246-2272
(503) 246-2282 (fax)
Download IConnect(R) TDR software demo at www.tdasystems.com/download!
----- Original Message -----
From: "istvan novak" <istvan.novak@xxxxxxxxxxxxxxxx>
To: <jan.vercammen.jv1@xxxxxxxxxxxxxxxx>; <si-list@xxxxxxxxxxxxx>
Sent: Wednesday, October 30, 2002 5:46 AM
Subject: [SI-LIST] Re: spatial resolution and effective rise time of VNA
with TDR time-domain option
>
> Hi Jan,
>
> My 2 cents:
> As you noted, the spatial resolution is different for different Dk media,
> so probably it is more effective to stay with bandwidth/risetime. The
> complications there are that VNAs
> - do not work all the way down to DC
> - usually have a range of windowing options for IFFT
>
> The fact that VNAs dont include DC in their frequency range is
> less of a problem when you want to resolve closely spaced
> discontinuities, but it creates an uncertainty of baseline if you
> have to do post processing.
>
> Windowing options may include flat, Hamming-like and Bessel-like
> windows. Their leakage is different, and it will directly affect the
> spatial resolution.
>
> With a 'typical' windowing, the first approximation of BW=0.35/tr
> is a good start, and you could substitute the upper frequency end
> of the sweep to BW, and get the equivalent tr.
>
> The benefit of using VNAs for TDR is that you can do a full
> calibration up to the probe tips, if you want to, otherwise it is
> usually more limited in resolution compared to true TDR.
>
> regards
>
> Istvan novak
> SUN Microsystems
>
>
> ----- Original Message -----
> From: "Jan Vercammen" <jan.vercammen.jv1@xxxxxxxxxxxxxxxx>
> To: <si-list@xxxxxxxxxxxxx>
> Sent: Wednesday, October 30, 2002 7:57 AM
> Subject: [SI-LIST] spatial resolution and effective rise time of VNA with
> TDR time-domain option
>
>
> >
> > Hello SI-list,
> >
> > I have a question about the saptial resolution of vector network
analysers
> > (VNA)
> > with time-domain TDR option, that is, to resolve discontinuitites in
> > space (=length dimension)
> >
> > Is there a rule of thumb for estimating the spatial resolution of a VNA
> > with a
> > TDR time-domain option from knowlegde of its bandwith BW.
> >
> > I have found one rule in an application note: 15cm/BW(Ghz). For example,
> > for a 6GHz
> > BW the spatial resolution is about 2.5cm for an effective dielectric
> > (dkeff) of 1.0.
> > You probably get better resolution for PCB or cable measurments. The
rule
> > of thumb would become: 15cm/(sqrt(dkeff) BW(GHz).
> >
> > Another question concerns the relation between the VNA bandwith and the
> > available
> > effective rise time of the step or pulse of the time domain signals. I
> > assume that one
> > cannot use the rule trise=0.35/Bw because there BW refers to the 3dB
> > breakpoint
> > (the effective BW is bigger). A VNA bandwith is its upper limit.
> >
> > -1- Is the above spatial resolution rule correct?
> > -2- Is there a rule of thumb to convert bandwith (BW) into an effective
> > rise time (10-90%)
> > of a VNA with a time-domain option?
> >
> >
> >
> >
> > Kind regards,
> >
> > Jan Vercammen
> >
> >
> > ------------------------------------------------------------------
> > 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 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 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 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:
