[TowerTalk] Antenna Analyzers

[Jim Brown]

Yes. The comparison is between lab equipment costing tens/hundreds of 
thousands of dollars and a very decent ham-produced unit that we can 
currently buy for less than $600.

And I've had no difficulty calibrating my VNWA 3e for 2-port 
measurements to several different measurement planes.  One of those 
produced the data for this report. 
http://k9yc.com/BandpassFilterSurvey.pdf  Screen plots of all the 
measured data are here.
http://audiosystemsgroup.com/BandpassFilterData.htm

Most of the time, my VNWA 3e has a 3m cable attached to its output and a 
1m cable to it's input. The 3m cable is a convenient length to reach 
connectors for antennas entering my shack.  I have several master 
calibrations for this measurement plane, and several more for a special 
test jig that W6GJB built for me to measure chokes. That test jig places 
the choke in series between input and output, S21 is measured, and the 
math function in VNWA software solves the voltage divider equation and 
plots Zmag, Rs, and Xs vs frequency.

73, Jim K9YC

On 12/5/2018 11:34 AM, Wes Stewart wrote:
> Rick,
>
> With all due respect, I'm not terribly surprised with your comments, 
> considering that you worked for HP. When I worked for Hughes, I 
> remember several occasions where I took an airplane ride from Tucson 
> to our sister facility in Conoga Park, CA just to use their network 
> analyzer to measure something or the other.  It took me several years 
> to get a capital equipment proposal approved by upper management to 
> purchase an HP8510.  As I recall, with all of the ancillary equipment, 
> software, cal kits, etc, the cost was in the neighborhood of 
> $200,000.  And it was still buggy and did require support.
>
> The VNWA software may not be as user friendly as you would like, but 
> it's remarkably free.  As I said, it will run the FA-VA5 and I didn't 
> mention it but it will even run the old N2PK.  There are automatic 
> selection options for the clock multipliers.  I suspect that a request 
> to Tom, might get an option to move the center frequency to the marker 
> written into a revision.  (I'm not speaking for him.)
>
> Wes  N7WS
>
> On 12/5/2018 9:13 AM, Richard (Rick) Karlquist wrote:
>>
>>
>> On 12/5/2018 7:21 AM, Wes Stewart wrote:
>>
>>> 2) Short of an HP-8510, 8753 or 3577 the gold standard for hobbyist 
>>> use IMHO is the SDR-Kits, DG8SAQ designed VNWA. 
>>> (https://www.sdr-kits.net/introducing-DG8SAQ-VNWA3) This is also a 
>>> two-port vector analyzer. The VNWA software is *very* powerful, with 
>>> a commensurate learning curve, and is well supported by the author 
>>> Dr.Tom Baier as well as some very knowledgeable users.  A bit 
>>> difficult to operate in the field because it's tethered to a 
>>> computer, but I have done it.  This is a lab grade instrument. Uses 
>>> SMA connectors.
>>
>> I would somewhat argue about the software being very powerful.
>> Especially compared the the HP analyzers you mentioned.  It
>> is not even close in terms of user productivity.  For example,
>> AFAIK, there is no command to set the center frequency to the
>> marker frequency.  The user would have to make a note of the
>> marker frequency and then manually enter the new center
>> frequency.  Just one of many limitations.  The user is also expected
>> to set the multiplication factors for RF and LO depending on the
>> measurement frequency.  This is very poorly documented.
>>
>> It is not clear if you can do full 2 port calibration.  This
>> is probably covered under the learning curve comment.  If it
>> is possible, it isn't obvious or straightforward.
>>
>> It is good that the VNWA is well supported but the HP analyzers you
>> cited are intuitively obvious to use and don't require any
>> support in most cases.
>>
>> (I have owned a VNWA for a few years).
>>
>> Rick N6RK
>>
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk