[Elecraft] K3 (not S) Antenna tuner strangeness

brian alsopb at comcast.net
Mon Jul 25 09:30:15 EDT 2016


Hi Don,

Further insight needed.

John and I have looked into the 80M tuner strangeness problem further.
We both have qualitatively seen the same thing.  It is related to trying 
to tune Z's that are on the order of 25-30 ohms on 160M/80M.

John already addressed the grounding issue in his K3.

The original problem raised was trying to tune a vertical antenna on 80M 
which had a measured R=30 ohms and XL =18 ohms.  The tuner only slightly 
reduced the SWR.  Repeated presses of ATU Tune did not improve things.

A calculation of the L network component values needed to match this 
load shows L/C values well within the range that the KAT could switch in.

I then tried matching a 25 ohm pure resistive load-- two 50 ohm dummy 
loads in parallel.  One would think this kind of load could easily be 
matched. Not so. It would match only on the upper end of 80M and not at 
all on 160M.  The tuner did not reduce the 2:1 SWR at all on 160M and 
only brought it down a few tenths on the low end of 80.

It is interesting to note that the high end of 80 match used L/C values 
greater than needed in the vertical antenna to match it.

John tried other resistive and combined resistive/reactive loads.  Here 
are his results.

Freq     Load value measured     SWR on K3        SWR on K3 after KAT3 tune
1.836        23.5 (via 1 m. RG213)     1.6    1.6
1.836        32+j28                             2.5 1.0
3.522        23.5                                 1.6 1.3
3.522        56+j55                             3.0 1.0
7.006        23.5                                 1.5 1.0
7.006        132-j76                            4.0 1.0
10.100       23.5                                1.3 1.0
10.100       30-j48                             4.0 1.0

So we have two different K3's having the same qualitative matching 
problem with lower Z loads on 80/160.

Of course it is possible that both KAT tuners have other component failures.

It would be interesting if others would try the two paralleled dummy 
load experiment.

Elecraft suggested shorting out the 8.215 MHz trap in series with the 
ATU/output.  They say the trap has been removed in later KAT versions.
The trap consists of parallel 96 nH and 3900 pf components.

It is not clear that would fix this particular issue.  It undoubtedly 
was a fix for other issues.

Looking at the component values and parallel configuration:
1) Both shorted would be the Elecraft fix
2) Open C: Parallel inductor would shunt the RF around the failed cap. 
On 80/160 the added series reactance would be <2 ohms.
3) Open inductor: Parallel C would introduce a series reactance of less 
than an ohm.
4) Both open, the tuner would not match on any band.

So we are left with a mystery.

Why are lower Z (~25-30 ohm) loads so hard to match on 80/160?

It would be instructive if others have luck matching the 25 ohm 
paralleled dummy load case on 80/160.

Insight in where else to look would be welcomed.

73 de Brian/K3KO


On 7/25/2016 2:35 AM, Don Wilhelm wrote:
> John,
>
> Yes, the "earthing" connection to the SO-239 jacks is important.
> That connection is provided by the mechanical connection to the standoff
> between the right side panel and the KAT3.
> The assembly manual instructions will provide adequate connection -
> first insert the side panel screw through the 2-D connector and then
> tighten the standoff to the 2D connector, and lastly, put the screw and
> lockwasher into the standoff.
>
> The problem is that *if* that long screw in the upper rear corner of the
> right side panel is removed for any reason, it cannot be tightened
> properly without going through the entire procedure - tighten the 2D
> connector, then tighten the standoff and finally tighten the screw into
> the standoff.
>
> If one tries to remove that screw from the right side panel, it is
> impossible to re-tighten everything without going through the entire
> process in order.  The threads of the 2D connector cannot be properly
> tightened to the threads of the standoff.
>
> A shortcut is to drill out the screw threads on that 2D connector
> allowing the screw to pass through the 2D connector without threads.
> That will permit the screw threads to pass through the 2D connector and
> be tightened onto the standoff.
>
> I recommend that anyone removing the right side panel from their K3/K3S
> to drill out the threads in this hole of the 2D connector so it can be
> assembled tightly without going through the entire procedure of first
> tightening the 2D connector, then the standoff and finally the screw and
> lockwasher.
>
> 73,
> Don W3FPR
>
> On 7/23/2016 9:09 PM, g3pqa at onetel.com wrote:
>> Guys,
>> I've run into something on 80M trying to use the internal antenna tuner.
>> It fails to tune to a low SWR. Brian K3KOI had a similar problem but
>> never quite got to the bottom of it because partly intermittent,
>> possibly multiple faults.
>> In the end to fix it:-
>> a) I rewound one of the swr sensor toroids (centre wire had possibly
>> arced to ferrite because I disconnected antenna by mistake and tuned),
>> and
>> b) fixed a new earthing wire between KAT3 chassis and SO259 Ant1
>> socket body. (possible poor earth connection via KAT3 securing pillar
>> to side panel)
>> John G3PQA
>>
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Elecraft at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to alsopb at comcast.net


More information about the Elecraft mailing list