FreeLists / si-list / [SI-LIST] Re: TDR vs. VNA? Which to purchase?

["McCoy, Bart" <McCoy.Bart@xxxxxxxx>]

Just a few comments, for what it's worth:

I had a chance to do a comparison:  TDNA solution vs VNA solution to
obtain 4-port S-param characterization to 10 GHz.   We were interested
in finding a simplified calibration solution to extract S-parameters.
As you know, VNA's are unforgiving when it comes to calibration while
TDNA seems to allow incremental accuracy tradeoffs with incremental
effort.  That was appealing.

What interested me in TDNA is (1) cheaper equipment  (2) Claims that
calibration was easier, requiring fewer standards that did not require
the same kind of characterization that VNA cal substrates  (I was doing
in-fixture probing/testing, not cable-end testing).

I compared an Agilent PNA (20 GHz model) to a Tektronix TDS8200 to
extract 4-port S-parameters. =20
My first test DUT was a 3 inch differential stripline net that had via
resonances... with a deep resonant loss (S21 < -40 dB) at resonance near
5 GHz. =20

VNA SOLT calibrations gave me repeatable results using 2 different probe
sets (GGB vs Cascade Microtech) and their corresponding cal. substrates.
Results agreed well, even over multiple calibration/measurement cycles
with the different probes.   So I trusted the data.

I repeated the same net measurement with TDNA and the results were
somewhat different.   Using only insertion thrus and an open reference,
the TDNA system produced excellent results.... up to 4 GHz.   Above 4
GHz, the results did not compare well.... 8-10 dB deviations from the
VNA/SOLT solutions that appeared random.  But not bad, considering I
only used an insertion/open reference standards.

We switched gears and performed cable test with a flex circuit and
switched to an Anritsu VNA.   This time, we did more variations in TDNA
calibrations.  As you know, TDNA calibration doesn't *require* a full
S-O-L-T calibration and we mixed different types:  shorts/thru cal,
open/thru cal, etc.   Regardless of the combination, we were unable to
replicate the VNA performance to 10 GHz.  Even the SOLT cal didn't match
the VNA (which we trusted more, given it was a cable cal).  Low freq
results matched within 0.75-1.0 dB up to 5-6 GHz.  Above that, results
generally did not agree, no matter what cal standards I threw at it. =20
The TDNA calibration combinations which did not use an insertion thru in
the measurement obviously looked worse when comaring S21... with lots of
ripple from low freq to 10 GHz.   The ones that did use an insertion
thru still showed several dB difference (TDNA compared to VNA) above 6
GHz. =20

Overall, if you're staying at 4-5 GHz or less, you may be okay with
TDNA, but I personally tend to trust it  less than the TDNA promo papers
do above 5 GHz.  I admit I'm no TDNA expert, but we gave it a good shot
in the lab and weren't thrilled at 5-10 GHz.  =20

Of course, the I-connect software has lots of bells and whistles you
won't get with PLTS, so it depends on what you want to measure.   But to
Agilent's credit, we compared a few dozen Tektronix differential TDR
measurements to Agilent TDD11/TDD22 curves (equivelent to differential
TDR, but calculated from S-parameters in the Agilent PLTS software).
The S-parameter based TDR compared VERY well to the Tektronix true time
domain differential TDR.   But we also took great care with the
Tektronix to do proper differential deskew according to the manual,
which is a must. =20

Overall, in terms of calibration ease... in any system... I think (based
on this experiment) you get what you "pay" for in terms of effort and
work calibrating (and you'll pay, one way or another).  If you shortcut
the cal, don't expect equivalent results.  Skip the S, O or L and you
get poor return loss at some point or other and probably some mismatch
loss as well.  Skip the T (thru) and you'll pay in your S21 for sure.
If you use all 4 S-O-L-T, then you'll do better, but it's still VERY
hard to beat (or even match) a VNA for S-parameters. =20

Then again, you may be looking for other capabilities beyond
S-parameters.

Good luck!

  - Bart





=20
Bart McCoy                          Email:  McCoy.Bart@xxxxxxxx =09
Mayo Foundation=20
4001 41st Street NW=20
MSC Sn 2-107                        Phone:  (507) 538-5465
Rochester, MN 55901                 Fax:    (507) 284-9171

"Fooling around with alternating current is just a waste of time. Nobody
will use it, ever."

Thomas Edison;  1889


-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
On Behalf Of Tom Cipollone
Sent: Tuesday, February 19, 2008 8:47 AM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] TDR vs. VNA? Which to purchase?

Hi,
  =20
  I realize that this thread has been done before, but as we all know,
technology changes pretty fast, and what might have been true a year or
two ago, may not still be true today.
  =20
  I am ready to make a large investment in signal integrity test
equipment for my company. I'm probably no more than a week away from the
decision. As some of you will no doubt suggest to get both, let me just
say that I can not afford both. I will have one or the other.
  =20
  I have had the demos and seen the equipment.=20
  =20
  What I do for a living is design boards for digital video, that are
comprised of many differential pairs, routed closely together. Rising
edges are in the neighborhood of 100 ps. I want to be able to find and
modify discontinuities (either capacitive or inductive), determine
crosstalk between the pairs and measusre jitter (yes, I will need a
signal source for that).
  =20
  If this were as little as two years ago I would have believed that the
VNA solution would have been the most versatile. However, Tektronix has
made a lot of progress with their TDR equipment and with their
"Iconnect" software. Also, there is a big difference in price between
the TDR solution and the VNA solution.
  =20
  In appealing to the SI group for opinions I am trying to go beyond my
own predjudice and the marketing hype.
  =20
  Thank You
  Tom
  =20

------------------------------------------------------------------
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:
http://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:    =20
                http://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
 =20
------------------------------------------------------------------
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:
http://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:     
                http://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