Towertalk
[Top] [All Lists]

Re: [TowerTalk] Baluns/tutorial/notes.

To: towertalk@contesting.com
Subject: Re: [TowerTalk] Baluns/tutorial/notes.
From: Ian White GM3SEK <gm3sek@ifwtech.co.uk>
Reply-to: Ian White GM3SEK <gm3sek@ifwtech.co.uk>
Date: Mon, 24 May 2010 09:20:23 +0100
List-post: <towertalk@contesting.com">mailto:towertalk@contesting.com>
Richard (Rick) Karlquist wrote:
>I meant to say "S21" not "S22"
>
>Rick N6RK
>
>Richard (Rick) Karlquist wrote:
>> Let me greatly simplify this discussion.
>>
>> Get a TWO port network analyzer and connect
>> the DUT from the center conductor of port 1
>> to the center conductor of port 2.  Measure
>> S22 and calculate the impedance of the DUT.
>> This eliminates most of the well known gotcha's
>> associated with measuring high impedances.
>> You should be able to measure even 50,000
>> ohms with no trouble.
>>
>> If you just want to know the magnitude of the
>> impedance, you can fake a network analyzer
>> by using a signal generator and a receiver.
>> Just measure insertion loss of the balun.
>>
>> Only works for floating DUT's but that includes
>> baluns.  It also doesn't matter much if the
>> terminals are far apart physically.
>>

For impedance measurements on chokes, I use the same physical layout 
that Rick describes above. The only differences are in the calibration 
procedure and the mathematical processing of the raw data.

Fundamentally, all VNA measurements are based on the attenuation and 
phase shift that is measured between the instrument's "TX" and "RX" 
ports. Depending on the type test fixture attached, and the type of DUT, 
the raw attenuation data can be processed in many different ways.

The test fixture that Rick describes can obviously be used for 
straightforward attenuation measurements; but less obviously it can also 
be considered as a 'bridge' [1] and the attenuation data can be 
processed in a different way to give impedance results. An appropriate 
calibration procedure must also be used, of course, but the rest of the 
processing is pure computation.

The important points are that the physical limitations of the test 
fixture due to its stray inductance and capacitance remain exactly the 
same; and above all, the need to prove the entire test setup by using a 
test load of the order of 10k.



[1] The use of unconventional 'bridges' to measure very high or very low 
impedances is covered in the design notes for the N2PK VNA:
http://n2pk.com/VNA/n2pk_vna_pt_2_ver_b2.pdf  starting at page 20.

It isn't at all obvious that the test fixtures in Figures 12 and 13 can 
be used as bridges, but on closer examination the math does work out 
correctly. The most appropriate type of test fixture is the one that 
gives the best S/N ratio for the measurement in question, and requires 
the smallest calibration corrections.





-- 

73 from Ian GM3SEK
http://www.ifwtech.co.uk/g3sek
_______________________________________________



_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk

<Prev in Thread] Current Thread [Next in Thread>