[Elecraft] KAT3 antenna tuner

[Kok Chen]

On Jan 31, 2010, at 1/31    10:34 AM, Roger D Johnson wrote:

> I started on this topic because the tuner failed to find a good  
> match at
> 1800.

To see the approximate range of the internal K3 antenna tuner at 1800  
kc, take a look here

http://homepage.mac.com/chen/Technical/Tuner/tuner1.8.pdf

(warning, it is a very large 9 MB PDF file (lots of points), if you  
download as PDF and view with a PDF reader, you can rescale, etc  
faster than viewing from a web browser).

Each dot indicates the impedance at the antenna port when the radio  
port of the tuner is set to 50 ohms.  So the tuner should tune an  
antenna that is the conjugate of where a dot lands on the Smith Chart.

The green points are for the case trx > series L : shunt C > antenna,  
and the blue points are when the tuner switches to trx > shunt C >  
series L > antenna.

> My VNA shows the impedance at the TX connector as 22+j8 which should
> be well within the proper range.


The conjugate of 22 + j8 ohms appears as 0.44 - j0.16 ohm on the  
normalized Smith Chart in the PDF file.  That is just barely inside  
the coverage region of the tuner.

You might try manually setting the tuner to the trx > shunt C > series  
L > antenna position, turn C to the max (1200+600+300+...) and then  
manually adjusting around the small values of L for lowest SWR.

 From my fiddling with the values, a 50 ohm source appears to prefer  
large shunt C followed by a small series L on the way to a 22 + j8  
ohms antenna .  It could be possible that other circuitry is  
preventing you from reaching precisely that area of the tuner.

One thing you can do is to move the impedance of what you see at the  
end of your current transmission line.  Since you have a VNA, you  
might want to play with changing the line length of the 50 ohm  
transmission line (or use a short length of a different impedance line  
to transform the impedance that you are currently seeing) so that its  
conjugate is more inside the dotted region of the PDF file.

Also, since you have a VNA, why not temporarily disconnect the  
transceiver output from the tuner, replace it with a 50 ohm resistor  
and look at antenna port of the tuner with the VNA?  You can then see  
the conjugate of the impedances which your tuner can transform into 50  
ohms.

73
Chen, W7AY

P.S.  Anyone who wants to run this program for other frequencies just  
send me a note and I can either send you the app or the Xcode project.  
You need to be running at least Mac OS X 10.5 (also works fine on  
10.6).  If you want to modify it for the KAT100 L and C values, you  
will need to change the component tables in the source, and recompile  
under the Xcode IDE.

Some caveats: the chart shows the combinations of C and L values  
without taking into consideration any strays.  It also does not take  
into consideration the effects of the KBPF3.